(Title of the Thesis)*

THE IDENTIFICATION OF BRCA1 AND BRCA2 MUTATION CARRIERS USING FUNCTIONAL GENOMIC ASSAYS

Claire S. Michel

A thesis submitted to the Department of Pathology & Molecular Medicine

In conformity with the requirements for

the degree of Master of Science

Queen’s University

Kingston, Ontario, Canada

March, 2008

An estimated 5-10% of breast cancers are hereditary in nature and are due to the presence of a mutation in a breast cancer predisposition gene; approximately half of these cases possess a mutation in BRCA1 or BRCA2. Many BRCA1/BRCA2 mutations result in a truncated protein and hence are unequivocally disease-causing. However another class of mutations, the Variants of

Unknown Significance (VUS), are more problematic as the effect of these mutations on protein function is unclear. The inability to classify these mutations as disease causing generates significant problems in risk evaluation, counseling and preventive care. Accordingly we sought to determine whether carriers of either a BRCA1 or BRCA2 mutation could be identified from non- carriers based on the gene expression patterns of non-cancerous cells.

EBV-transformed lymphoblastoid cell lines established from BRCA1/BRCA2 mutation carriers and normal individuals were obtained through the NIH Breast Cancer Family Registries.

Cell lines were mock-irradiated or treated with ionizing radiation (2 Gy). Following a recovery period of 6 hours total RNA was extracted and whole genome gene expression profiling was carried out. Molecular classifiers comparing the baseline expression profiles and the radiation- dependent expression profiles of BRCA1/BRCA2 mutation carriers to control individuals were created using a Support Vector Machine (SVM) coupled with a recursive feature removal (RFR) algorithm.

Our results suggest that cell populations derived from BRCA1/BRCA2 mutation carriers display unique expression phenotypes from those of control individuals in both the basal and radiation-induced cases. In the task of classification using baseline expression, the BRCA1- classifier correctly classified 15/18 test samples using feature selection based on the training set only, while feature selection using the entire dataset (AD) improved classification to 16/18 samples. The BRCA2-baseline classifier correctly classified 13/17 and 14/17 (AD) samples, ii respectively. In the task of radiation-dependent classification, the BRCA1-IR classifier correctly classified 12/18 and 16/18 (AD) test samples respectively while the BRCA2-IR classifier correctly classified 13/17 and 16/17 (AD) test samples respectively. These results suggest the possibility of development of this assay into a novel hereditary breast cancer screening diagnostic able to accurately identify the presence of BRCA1 or BRCA2 mutations via a functional assay thereby improving patient outcomes.

iii

Acknowledgements

Firstly I would like to thank my supervisors, Dr. Scott Davey and Dr. Harriet Feilotter for continually challenging and supporting me over the course of my work. To the members of the

Davey lab past and present: it is always a pleasure.

I would like to thank my family; Jane, Dave and Alex for always encouraging me to strive to do well in whatever I pursue, and for letting me believe the sky is the limit. Mum, thank you for always thinking of us: I dedicate this work to you.

I would like to thank ―The Family‖: Jenny, Rob, Catriona, Pou and especially Larbi, you guys mean to world to me and I would be only a shadow of the person I am today without your constant love, encouragement and understanding. To be surrounded by such loving and caring people is a truly wonderful gift. Larbi, thank you for always encouraging me in my endeavors—to know that I can always count on you is beyond words.

To the Science Geeks: Matt, Cheryl, Joe, Jess, Shawna, Andrea, Jalna, Mia, Morgan, Dr. C & The

Crudden Lab: thank you for many hours of entertainment, enrichment-ridiculousness, white board races and much, much, more.

Lastly, to the Queen’s Triathlon & Cycling Teams: to have been fortunate enough to spend so many hours around such hard-working, motivated, kind people was truly my honour.

Table of Contents

Abstract ...... ii Acknowledgements ...... iv Table of Contents ...... v List of Figures ...... viii List of Tables ...... ix List of Abbreviations ...... x

Chapter 1 Introduction & Literature Review 1.1 Breast Cancer 1.1.1 The Disease At Large ...... 1 1.1.2 Hereditary Breast Cancer ...... 1 1.1.3 BRCA1 Structure & Its Relationship to Breast Cancer ...... 3 1.1.4 BRCA2 Structure & Its Relationship to Breast Cancer ...... 5 1.1.5 Classical Tumor Suppressor Genes or Otherwise? ...... 5 1.2 The Cell Cycle & Control via Checkpoint Activation 1.2.1 The Cell Cycle ...... 6 1.2.2 Cell Cycle Checkpoints Maintain Genomic Integrity ...... 7 1.2.3 The Checkpoint Proteins ...... 7 1.3 The Cell Cycle Checkpoints 1.3.1 The G1/S-Phase Checkpoint ...... 11 1.3.2 The G2/M-Phase Checkpoint ...... 12 1.3.3. The Functions of BRCA1 in the Cell Cycle Checkpoints...... 12 1.4 DNA Damage Response 1.4.1 DNA Damaging Agents ...... 13 1.4.2 DNA Double Stranded Breaks (DSBs) ...... 14 1.4.3 DNA Damage Repair ...... 15 1.5 DNA DSB Repair Pathways 1.5.1 NHEJ ...... 15 1.5.2 HR ...... 17 1.5.3 SSA ...... 20 v

1.6 BRCA1 & BRCA2 Function in the DNA Damage Response ...... 20 1.6.1 BRCA1 in DNA Damage Repair ...... 21 1.6.2 BRCA2 in DNA Damage Repair ...... 22 1.7 Microarrays ...... 23 1.7.1 The Components Of A Microarray ...... 23 1.7.2 Normalization ...... 25 1.7.3 Hypothesis Testing...... 25 1.8 Classification Methods 1.8.1 Unsupervised Classification ...... 26 1.8.2 Supervised Classification ...... 27 1.8.3 Recursive Feature Removal ...... 28 1.9 Hypothesis & Objectives ...... 30

Chapter 2 Materials & Methods 2.1 Cell Culture ...... 31 2.2 DNA Damage Induction ...... 31 2.3 Cell Cycle Experiments 2.3.1 Determination of IR treatment dose and Recovery Time ...... 31 2.3.2 G1/S-Phase Checkpoint Activation: Visualization via Flow Cytometry .... 33 2.4 Gene Expression Profiling ...... 34 2.5 Data Analysis 2.5.1 Filtering & Normalization ...... 35 2.5.2 Classification Algorithms ...... 36 2.5.3 Visual Representation of Data ...... 40

Chapter 3 Results 3.1 Cell Cycle Checkpoint Activation is Present in all Genotypes ...... 41 3.1.1 Determination of Optimal IR Treatment Dose ...... 41 3.1.2 Elucidation of Optimal DNA Damage Recovery Period ...... 41 3.2 The DNA Damage Response is Present in Wild Type (Control) Cells ...... 42

3.3 BRCA1/BRCA2 Mutation Carriers are Readily Identifiable at the Gene-Expression Profiling Level Following the Induction of DNA Damage ...... 46 3.3.1 Analysis I: Construction of BRCA1/BRCA2 Radiation-Dependent Classifiers with Independent Validation ...... 49 3.3.2 Analysis II: Construction of BRCA1/BRCA2 Radiation-Dependent Classifiers Using the Entire Dataset for Feature Selection ...... 51 3.3.3 Comparison of Features Selected in Analyses I & II ...... 52 3.4 BRCA1/BRCA2 Mutation Carriers Are Readily Identifiable at the Baseline Gene Expression Profiling Level ...... 54 3.4.1 Analysis I: Construction of BRCA1/BRCA2 Baseline Classifiers with Independent Validation ...... 54 3.4.2 Analysis II: Construction of BRCA1/BRCA2 Baseline Classifiers Using the Entire Dataset for Feature Selection ...... 62 3.4.3 Comparison of Features Selected in Analysis I & II...... 62 3.5. Feature Selection for Classification vs. Biological Insight ...... 63

Chapter 4 Discussion 4.1 General Discussion ...... 65 4.2 IR-Induced Response in WT Cells...... 65 4.3 SVM: An Efficient Tool for Microarray Analysis ...... 67 4.4 Differences in gene expression observed between BRCA1/BRCA2 mutation carriers and wild type individuals at the baseline expression-profiling level are likely due to Inefficiency in Endogenous DNA Damage Repair ...... 67 4.5 Differences in gene expression observed between BRCA1/BRCA2 mutation carriers and wild type individuals at the radiation-dependent expression-profiling level are likely due to inefficiency in the repair of exogenous DNA Damage ...... 71 4.6 Future Directions ...... 76 4.7 Clinical Relevance ...... 77

References ...... 79 Appendix A: Supplementary Tables ...... 91

vii

List of Figures

Chapter 1:

Figure 1: Schematic diagram of the BRCA1 & BRCA2 Proteins ...... 4

Figure 2: Cell cycle checkpoint control in response to DNA damage ...... 10

Figure 3: The major DNA DSB repair pathways: NHEJ & HR ...... 19

Figure 4: A pictorial representation of an SVM ...... 29

Chapter 2:

Figure 5: A working example of the RFR algorithm ...... 39

Chapter 3:

Figure 6: Flow cytometry histograms: visualization of checkpoint induction ...... 43

Figure 7: Determination of optimal recovery period ...... 44

Figure 8: Flow chart outlining step-wise construction of molecular classifiers ...... 48

Figure 9: Hierarchical clustering of features comprising the non-radiation dependent (baseline) BRCA1 classifier ...... 57

Figure 10: Hierarchical clustering of features comprising the non-radiation dependent (baseline) BRCA2 classifier ...... 59

Figure 11: Gene Ontology categories of features comprising the BRCA1 & BRCA2 non- radiation dependent classifiers ...... 61

viii

List of Tables

Table 1: Specific BRCA1 & BRCA2 mutations used in this study ...... 32

Table 2: Genes differentially expressed in wild type cells following treatment with IR: the wild type IR-response ...... 45

Table 3: BRCA1/BRCA2 IR-dependent classification: Analyses I & II results ...... 50

Table 4: Comparison of IR-dependent features identified in Analysis I vs. Analysis II for BRCA1 & BRCA2 ...... 53

Table 5: BRCA1/BRCA2 non radiation-dependent (baseline) classification: Analyses I & II Results ...... 56

Table 6: Comparison of non-IR (baseline) features identified in Analyses I vs. Analysis II for BRCA1 & BRCA2 ...... 64

List of Abbreviations

BRCT Breast Cancer Associated Gene 1 Carboxy-Terminal NHEJ Non-Homologous End Joining ATM Ataxia Telangiectasia Mutated ATR Ataxia Telangiectasia Mutated and Rad3 Related HR Homologous Recombination G1 Gap1 G2 Gap2 M Mitosis S Synthesis Phase DSB Double Stranded Break MRN Mre11-Rad50-Nbs1 PBS Phosphate-Buffered Saline SSB Single Stranded Break Cy3 Cyanine 3 (Green Fluor) Cy5 Cyanine 5 (Red Fluor) BrdU Bromodeoxyuridine PI Propidium Iodide IR Ionizing Radiation cDNA Copy DNA cRNA Copy RNA SVM Support Vector Machine EBV Epstein Barr Virus LCL Lyphoblastoid Cell Line RNA Pol II RNA Polymerase II ssDNA Single-Stranded DNA dsDNA Double-Stranded DNA SSA Single Strand Annealing UV Ultraviolet WT Wild type DNA Deoxyribonucleic acid

Chapter 1: Introduction & Literature Review

1.1 Breast Cancer:

1.1.1 The Disease At Large

Currently in Canada, breast cancer accounts for the second highest incidence rate and is the second leading cause of cancer-related death in women. According to the National Cancer

Institute of Canada in 2007 the disease was diagnosed in an estimated 22,300 women, with 5,300 women suffering breast-cancer related deaths [1]. Factors contributing to incidence of the disease include a number of known genetic and environmental predispositions. While approximately 90% of breast cancers occur sporadically, without known predisposing genetic alterations, the remaining cases are linked to heritable causes which are known to include mutations in tumor suppressor genes, including most notably, though not exclusively, BRCA1 and BRCA2 [2]. A common theory on the evolution of human breast tumors is that they arise as a result of multiple genetic changes which gradually transform differentiated and growth-limited cells into highly invasive cells which are no longer responsive to cellular growth controls [3]. The genetic evolution of normal breast cells into cancer cells is thus largely determined by the fidelity of DNA replication and repair [4]. Overall, the complex etiology of breast cancer hampers its prevention, diagnosis, and treatment alike.

1.1.2 Hereditary Breast Cancer:

Over one hundred years ago Paul Broca first described a hereditary breast cancer syndrome that is now recognized to account for approximately 5-10% of all breast cancer cases

[5]. Because the disease showed a high prevalence in certain families the notion that a subset of breast cancer cases could be linked to the inheritance of predisposing mutations in cancer-

1 susceptibility genes was put forth [6]. Key evidence that heritable breast cancer risk could be caused by a single genetic mutation came with the identification of a locus on chromosome 17q that was linked to disease susceptibility in specific families [7]. In 1994 the BRCA1 gene was subsequently identified through positional cloning [8]. As only 45% of familial breast cancers showed evidence of linkage to BRCA1, with a most notable absence of linkage in families suffering a high incidence of male breast cancer, the search for a second breast cancer susceptibility gene continued and in 1995 BRCA2 was identified at chromosome 13q12.3

[9,10,11].

The majority of BRCA1 and BRCA2 breast cancer susceptibility mutations (~ 70%) are known to result in a truncated, and therefore inactive, gene product. This class of mutations can be readily identified using currently available genetic screening methods, and are likely all clinically significant [12]. The remaining BRCA1 and BRCA2 cancer-predisposing mutations do not interrupt the open reading frame of the genes, but instead are found to be missense, or in- frame splicing mutations. The current number of mutations falling within this class cumulatively numbers over 1400 to date [12]. Unlike truncating mutations, establishing the clinical relevance of this subset of mutations remains a major challenge, and these mutations are hence termed

―Variants of Unknown Significance‖ (VUS). Current estimates place individuals who inherit a single germline mutation in either BRCA1 or BRCA2 at a lifetime risk of approximately 80% of developing breast or ovarian cancer by age 70 years [13]; however the reasons why these mutations contribute to the development of breast and ovarian cancer remain poorly understood.

Elucidation of the precise molecular mechanisms of these genes could potentially improve our understanding of the way in which these mutations contribute to the development of the disease as well as offer new options for diagnosis and treatment.

1.1.3 BRCA1 Structure & Its Relationship To Breast Cancer

The BRCA1 gene encodes a large 1863 amino acid protein with multiple functional domains (Fig.1). The well characterized BRCA1 domains consist of an N-terminal RING domain—a globular zinc-binding motif that has been implicated in ubiquitination pathways [14], and two C-terminal tandem repeat globular domains termed BRCT [15] now known to be a common structural feature of proteins involved in the DNA damage response [16]. Mounting evidence implicates BRCA1 in all phases of the cell cycle and in the regulation of orderly events during cell cycle progression. Consequently, BRCA1 deficiency results in defects in the G1/S- phase checkpoint (indirectly), the S-phase checkpoint, and the G2/M checkpoint [17,18].

BRCA1 interacts with a large number of diverse molecules including tumor suppressors, oncogenes, DNA damage repair proteins, cell cycle regulators, and transcriptional activators and repressors [17]. Consistent with this expansive pattern of interaction, loss-of-function mutations in BRCA1 result in pleiotrophic phenotypes including growth retardation, increased apoptosis, defective DNA damage repair, defective G2/M cell cycle checkpoint, chromosome damage, and aneuploidy (reviewed in [18]). BRCA1 cancer-predisposing truncating and missense mutations, validated through functional analysis, are commonly found within the two C-terminal BRCT motifs and to a lesser extent in the critical Zn2+ binding residues within the N-terminal RING finger, indicating that these regions are critical for tumour suppressor function [19]. It is proposed that mutations in BRCA1 do not directly result in tumor formation, but instead cause genetic instability thereby subjecting cells to a high risk of malignant transformation [20].

Figure 1: Schematic Diagram of BRCA1 & BRCA2 Proteins A) The BRCA1 protein: well characterized domains include the N-terminal RING finger, the centrally located Nuclear Localization Signal (NLS) and the two C-terminal BRCT domains. B) The BRCA2 protein: well characterized domains include the 8 BRC repeats located in the center of the protein and the N-terminal NLS

1.1.4 BRCA2 Structure & Its Relationship To Breast Cancer

The human BRCA2 gene encodes a 3418 amino acid protein that is one of the largest polypeptides in the human proteome (Fig.1) [11]. Well characterized domains of the BRCA2 protein include a repeated motif termed the BRC domain, located centrally, which consists of a series of eight repeated BRC sequences [21]. It is these BRC domains, in concert with the C- terminal region of the protein, that mediate direct binding of BRCA2 to Rad51. Of the eight BRC repeats, Rad51 has been shown to bind to six of them in vivo, showing the highest affinity for

BRC3,4 and very low affinity for BRC5,6 [22,23]. This interaction highlights the importance of the BRC domains in facilitating BRCA2’s role in DNA double-stranded break (DSB) repair as the physical interaction of BRCA2 and Rad51 is essential for error-free homologous recombination

(HR) to take place. Another domain important to BRCA2 function in HR is the C-terminal nuclear localization signal (NLS). It is thought that the NLS on BRCA2 facilitates the transport of

Rad51 into the nucleus to sites of DNA damage as a NLS has yet to be identified on Rad51, and in BRCA2 deficient cells nuclear transport of Rad51 is found to be impaired [23]. As in the case of BRCA1, the majority of known carcinogenic BRCA2 mutations result in a truncated and therefore non-functional protein [12]. Thus it is proposed that mutations in BRCA2 cause increased chromosomal aberrations and genetic instability through the loss of error-free DNA

DSB repair via HR thereby subjecting cells to a higher than normal risk of malignant transformation [24].

1.1.5 Classical Tumour Suppressor Genes or Otherwise?

BRCA1 and BRCA2 are generally thought of as classical tumor suppressors as tumourigenesis in individuals with germline BRCA mutations is frequently accompanied by 5

inactivation of the remaining wild-type allele [25,26], however it is also known that one defective copy of BRCA1/BRCA2 in the germline is enough to cause cancer predisposition [18]. Whether

BRCA1 and BRCA2 function as classic (two-hit) tumour suppressors, or whether they promote tumorigenesis in the haploinsufficient state remains controversial.

1.2 The Cell Cycle: Checkpoint Activation and Control

1.2.1 The Cell Cycle

Each time a eukaryotic cell divides its genome must first be precisely duplicated and then equally divided and distributed into two new daughter cells. The complex series of events that occur during this process comprise the eukaryotic cell cycle which is divided into four distinct stages. These stages include: Gap1(G1) the gap phase prior to DNA replication; synthesis (S) during which DNA is replicated; Gap2 (G2) the gap phase following DNA replication, and mitosis (M), during which chromosomes are segregated and cell division takes place [27]. Cell cycle progression is regulated by the activity of protein kinase complexes consisting of a cyclin and a cyclin-dependant kinase (Cdk) [28]. As implied in the name, Cdks are tightly regulated through their association with cyclin proteins. In mammals, growth factors stimulate the expression of D-cyclins that bind to, and activate Cdk4 and Cdk6 to promote entry into G1 from quiescence [29]. Cdk2 associates with cyclin E and cyclin A to initiate and promote DNA replication, whereas Cdc2 (Cdk1) requires association with cyclin B to promote mitosis [30].

1.2.2 Cell Cycle Checkpoints Maintain Genomic Integrity

A large number of proteins are responsible for maintenance of genome integrity and the functions of these proteins include DNA synthesis, DNA damage repair, and regulation of cell cycle checkpoints [27]. Cell cycle checkpoint is a term used to describe regulatory pathways that function to prevent the initiation of downstream cell cycle events prior to the successful completion of earlier events [27]. At critical transitions in the eukaryotic cell cycle signaling pathways monitor the successful completion of upstream events prior to proceeding forward to the next stage [27]. Checkpoints that specifically respond to DNA damage include the G1/S-phase checkpoint, the S-phase (or intra-S) checkpoint, and the G2/M checkpoint. Cells lacking functional checkpoints display genomic instability due to a failure to properly respond to DNA damage, faulty DNA replication, or aberrant chromosome segregation [31]. If cellular damage cannot be properly repaired cell cycle checkpoint signaling can lead to the activation of pathways initiating programmed cell death [32].

1.2.3 The Checkpoint Proteins

The proteins responsible for cell cycle checkpoint activity are often categorized into four major classes that act in concert to translate the signal of damaged DNA into cell cycle arrest and

DNA repair (Fig.2). These classes include (a) the DNA damage sensor proteins, whose action is to recognize damaged DNA thereby initiating a biochemical cascade of activity; (b) mediator proteins, which relay the damage signals received from the sensor proteins onward; (c) transducer proteins and (d) effector proteins. The transducers proteins are protein kinases and phosphatases that serve to relay the damage signals to their final targets the effectors, which are protein complexes that are directly responsible for transitioning between the phases of the cell cycle [33].

Proteins involved in the checkpoint signaling cascade can hold membership in more than one of these classes; that is to say, there is not an absolute delineation between the various components of the checkpoint. The DNA damage checkpoints are initiated in response to both endogenous and exogenous genotoxic agents which can result in a number of different lesions. Specific lesions activate specific branches of the checkpoint response, however like the protein classes, these responses are not absolute and often different classes of lesions trigger overlapping responses

[34]. The DNA damage sensor proteins are divided into two distinct groups based on structure: the phosphoinositide 3-kinase-like kinase (PIKK) family members ATM and ATR, and two protein complexes believed to function in concert: the 9-1-1 complex, a heterotrimeric ring containing Rad9, Hus1, and Rad1 and the Rad17-RFC complex [35,36]. ATM and ATR are both serine/threonine protein kinases which are each activated by the presence of specific DNA lesions. ATM activity is stimulated in vivo by the induction of DNA DSBs [37] while the ATR signaling cascade is activated by the presence of ssDNA regions [38]. Once active, the DNA damage sensors relay the message of damaged DNA by initiating a signaling cascade that starts with the group of proteins known as mediators. Proteins currently classified as mediators include p53 Binding Protein 1 (53BP1) [39], Topoisomerase Binding Protein 1 (TopBP1) [40] Mediator of DNA Damage Checkpoint Protein 1 (MDC1) [41], BRCA1 and Claspin [42]. The targets of these mediator proteins are the effector kinases, the best-characterized of which include the serine/theronine kinases Checkpoint kinase 1 (Chk1) and Checkpoint kinase 2 (Chk2) [43,44].

Chk2 is responsible for transducing the DSB signal relayed by ATM, while Chk1 is activated in response to the UV-damage signal sensed by ATR with some overlap in function being observed

[45, 46]. The ultimate targets of the cell cycle checkpoint response are the Cdk/cyclin protein complexes which directly regulate the cell cycle transitions [33].

Figure 2: Cell Cycle Checkpoint Control following the Induction of DNA Damage

Depending on the type of DNA damaging agent and the lesion it inflicts, different branches of the DNA damage response pathways are activated. The checkpoint proteins are loosely classified into four classes. Damaged DNA is recognized by the sensor proteins including ATM, the MRN complex, the 9-1-1 complex, Rad17-RFC, and ATR/ATRIP. The damage signal is then relayed to the mediator proteins which include TopBP1, Claspin, MDC1, 53BP1 and BRCA1 which in turn relay the damage signal to the transducers: Chk1 & Chk2. The transducer kinases and phosphatases then relay the signal onwards to the effector proteins, the Cdk/cyclin complexes which are directly responsible for transitioning between phases of the cell cycle.

1.3 The Cell Cycle Checkpoints

1.3.1 The G1/S-phase Checkpoint

Eukaryotic cells suffering insult in G1 exhibit a prominent delay prior to entry into S- phase. This arrest is vital to the maintenance of genome integrity as it allows the cell the time necessary to repair the inflicted damage thereby preventing the replication of a damaged template

[32]. The progression from G1 to S-phase is regulated by the activity of the cyclin D/Cdk4,6 and the cyclin E/A/Cdk2 complexes [32]. The G1/S checkpoint has dual phases of action; one to initiate and one to maintain the G1/S-phase arrest [47]. The first phase occurs within minutes of damage-induction and achieves G1-arrest through the post-translational modification of proteins in a p53 independent manner. Induction of DNA damage sensed by ATM activates Chk2 which phosphorylates Cdc25A thereby priming it for ubiquitination and proteasome destruction [48].

Degradation of Cdc25A maintains Cdk2 in an inactive, phosphorylated state, thereby inhibiting its interaction with cyclin E, a necessary interaction for progression into S-phase [48]. The second wave of action controlling the progression from G1 to S-phase occurs at the transcriptional level and therefore takes several hours post-damage to initiate. ATM/ATR and Chk1/Chk2 activate and stabilize p53 through phosphorylation at ser15 and ser20 respectively [49]. This phosphorylation prevents its nuclear export and degradation; the resulting increased levels of p53 cause transcriptional induction of p21, which in turn binds to, and inhibits, the Cdk2/cyclin E complex thereby maintaining G1/S-phase arrest [29,50]. p21 also binds the Cdk4/cyclin D complex preventing its phosphorylation of Rb [51]. Phosphorylation of Rb causes its dissociation from the

E2F family of transcription factors, whose actions are required for the transcription of S-phase genes [47].

1.3.2 The G2/M Checkpoint

The G2/M checkpoint ensures that cells harboring DNA damage do not undergo mitosis

[52]. Depending on the type of DNA lesion present, either the ATR/ATRIP-Chk1 signaling pathway, or the ATM-Chk2-Cdc25 pathway is activated [46]. The G2/M checkpoint is unique in that these two pathways are intertwined, with one initiating cell cycle arrest, while the other works to maintain it [34]. In the case of ionizing radiation (IR)-induced DSBs, the ATM-Chk2 signaling pathway initiates arrest, while the ATR/ATRIP-Chk1 pathway is responsible for its maintenance [34]. The common end-target of these pathways, directly controlling the transition through G2 to M, is the cyclin B/Cdc2 complex [53]. Control of the cyclin B/Cdc2 complex is achieved through down-regulation of Cdc25A and up-regulation of Wee1 [54]. Phosphorylation of Cdc25A by Chk1 and Chk2 causes its association with the 14-3-3 class of proteins; this association causes Cdc25A to become sequestered in the cytoplasm where it is tagged for degradation by the ubiquitin-proteasome pathway [55]. Loss of Cdc25A results in maintenance of the inhibitory phosphorylation of Cdc2 at Y15, and hence mitotic arrest. Concurrently, up- regulation of Wee1 in response to Chk1/Chk2 phosphorylation leads to mitotic arrest through its inhibitory phosphorylation of Cdc2 at Y15 and T14 [53].

1.3.3 The Functions of BRCA1 in the Cell Cycle Checkpoints

Recent findings suggest that BRCA1 is active in all phases of the cell cycle and plays an important role in coordinating cell cycle progression. BRCA1 has been shown to play numerous roles in the G1/S-phase checkpoint. Firstly, a study conducted by Somasundaram et al. [56] found that over-expression of BRCA1 stimulated transcription of p21 in a p53-independent manner and 12

prevented cell cycle progression into S-phase. As well, the BRCA1-BARD1 complex has been shown to be required for ATM -mediated phosphorylation of Chk2 and p53 at Ser15 which is required for G1/S-phase arrest via transcriptional induction of p21 [57]. Finally, BRCA1 was shown to interact with hypophosphorylated RB to inhibit cell proliferation and induce G1-arrest

[58]. Aprelikova et al. [58] suggested that BRCA1 binding to RB maintains it in a hypophosphorylated state which in turn maintains its interaction with E2F thereby preventing transcription of downstream genes required for S-phase progression.

Along with these roles in the activation and maintenance of G1/S-phase checkpoint,

BRCA1 is known to regulate the G2/M checkpoint at several levels. Firstly, it was recently found that the phosphorylation of BRCA1 by ATM at Ser1423 is required for activation of the G2/M checkpoint [59]. As well, evidence linking BRCA1 to the regulation of several key G2/M checkpoint effectors comes from a study conducted by Yarden et al. [54]. Yarden and colleagues found that BRCA1 regulates the expression, phosphorylation, and cellular localization of Chk1, a known regulator of the G2/M-phase checkpoint [54]. Their data also suggests that BRCA1 induces expression of both Wee1 and the 14-3-3 family of proteins [54], while a study conducted by Mullan et al. [61] demonstrated the ability of BRCA1 to induce expression of GADD45; all three are known to act as regulators of the Cdc2/cyclin B complex [60, 61].

1.4 DNA Damage Response

1.4.1 DNA Damaging Agents

Different genotoxic agents vary in the type of DNA damage they inflict and the specificity of the induced damage triggers a variety of cellular responses specific to the type of

lesion inflicted. While a large host of agents are known to activate checkpoint pathways, two commonly employed agents include IR and ultraviolent (UV) light. IR, by definition, is radiation with sufficient energy to ionize molecules with which it collides [62]. IR can damage DNA directly, or indirectly, through reactive oxygen species intermediates [62]. IR is known to induce a large variety of DNA lesions, the most lethal of which is the DNA Double Stranded Break

(DSB) [62]. Experimental doses of radiation commonly used range from 1-50 Gy, where one Gy of gamma radiation is hypothesized to cause 600-1000 SSBs, 16-40 DSBs, in the human genome

[63].

1.4.2 DNA Double Stranded Breaks (DSBs)

The most lethal form of DNA damage is generally regarded to be the DSB. DSBs are generated endogenously, as a normal part of the cellular process, through replication fork collapse, during DNA replication and in repair events, and by exogenous agents such as ionizing radiation (IR) and other genotoxic compounds [62]. Repair of DSBs is of cardinal importance in preventing chromosomal fragmentation, translocations and deletions. The genomic instability resulting from persistent or incorrectly repaired DSBs can lead to carcinogenesis through activation of oncogenes, inactivation of tumour-suppressor genes, or loss of heterozygosity

(LOH) at specific loci, while in the germline they can lead to inborn defects [64]. The deleterious effects of DSBs have triggered the evolution of multiple pathways for their repair [65].

1.4.3 DNA Damage Repair:

Mammalian cells possess three well-characterized pathways for DSB repair including

Non-homologous end joining (NHEJ), single-strand annealing (SSA), and HR [66]. NHEJ is favored by cells in G0/G1, and involves the direct re-ligation of the broken DNA ends; it is an inherently error prone process due to the loss of terminal bases that are frequently removed in order for ligation to occur efficiently [67]. NHEJ does not rely on extensive sequence recognition and thus has the capacity to ligate DNA ends from non-homologous chromosomes resulting in increased chromosomal aberrations [66,67]. SSA, also an inherently error prone process, promotes DSB repair by annealing short regions of sequence homology that border the DNA break site [6]. SSA results in the deletion of intervening sequences and thus loss of genetic information. HR, the most accurate of the three repair mechanisms, predominates over the other

DSB repair pathways during the S and G2 phases of the cell cycle in which sister chromatids are readily available [6]. It is initiated by nucleolytic processing of the DSB to generate single- stranded DNA (ssDNA) overhangs which act as substrates for the recombinase enzyme Rad51

[6]. Rad51 forms a nucleoprotein filament on the newly processed ssDNA overhangs, and then searches within the context of the nucleoprotein filament for an intact sister chromatid which acts as a template for DNA synthesis to repair the DSB [66,67]. The accurate repair of damaged DNA is essential for the maintenance of genomic integrity.

1.5The DNA DSB Repair Pathways

1.5.1 NHEJ

Cells in G0/G1 favor NHEJ over other repair pathways as the presence of small mutations in place of potentially lethal DSBs upon advancement into S-phase is far better tolerated by the 15

cell [68]. NHEJ uses no, or extremely limited, sequence homology to rejoin juxtaposed DNA ends in an error prone manner that frequently causes mutations and micro-deletions. As well, this repair pathway is thought to be one of the main causes of chromosomal rearrangements and translocations in the cell caused by the joining of previously unlinked DNA molecules [24].

Activation of the NHEJ repair pathway is thought to be dependent on the activity of the Mre11-

Rad50-Nbs1 (MRN) complex [69]. The MRN complex is one of the first complexes to sense

DSBs, and is localized to lesions within minutes of their formation [70]. Aside from its early role in the sensing DNA damage, relocation of MRN to the sites of DNA damage is thought to play a structural role by tethering together, and therefore stabilizing, broken chromosomes [71]. As well, in vitro studies have shown Mre11 to have dsDNA endo/exonuclease activity which is stimulated by the ATP binding actions of Rad50 [72]. DNA damage sensed by the MRN complex is transduced to the checkpoint effectors through interaction with ATM [71]. Following initiation of checkpoint arrest by MRN, the initial event in NHEJ involves the biding of the Ku70/Ku80 heterodimer to the broken DNA ends [73]. The two Ku-heterodimers must then come together to form a ring shaped structure to bridge matching DNA ends, the formation of which recruits the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs) [74]. The kinase activity of

DNA-PKcs is required for phosphorylation of target proteins whose endonuclease activity is required to produce compatible DNA ends at the break site as DSBs rarely result in compatible ends [75]. For example, IR generates nucleotides with aberrant structures such as 3'-phosphate or

3'-phosphoglycolate groups which must be removed by nuclease activity before ligation can occur

[76]. Following end-resectioning, compatible ends are created by the activity of DNA polymerase

λ (pol λ) after which the newly compatible ends are ligated by the XRCC4/DNA Ligase IV

complex (Fig.3) [73]. Inherent errors associated with NHEJ are avoided when cells employ an alternate DSB repair pathway: HR repair.

1.5.2 Homologous Recombination (HR):

HR requires extensive regions of DNA homology and repairs DSBs accurately using information on the undamaged sister chromatid [6]. HR can be carried out solely during the S and

G2-phases of the cell cycle, as sister chromatids are only available to act as templates for DNA repair during these times [77]. The major proteins involved in HR repair are BRCA1, BRCA2,

MRN, RPA, Rad51, Rad52, and the rest of the Rad52 epistasis group (Rad52b, Rad54, Rad54b)

[38,78]. Suggested roles of the MRN complex in HR repair include bridging and maintenance of proximity of the two DNA ends and nucleolytic resection of the DSB DNA ends, however roles in resection are questionable as MRN possess only 3’—5’ dsDNA exonuclease activity—the opposite polarity of that required [38, 71]. Following resection of the DSB ends the resulting 3’ overhangs become coated by Replication Protein Factor A (RPA), which acts to prevent and remove secondary structures [24]. Next, Rad52 is targeted to the newly processed ssDNA ends through interaction with RPA, which in turn promotes binding of Rad51 [77,78]. In order for

Rad51 to bind the ssDNA ends it must first be released from BRCA2, which is believed to be required for movement of Rad51 from its site of synthesis to the site of DNA damage [77,78].

Once free from BRCA2, Rad51 forms a nucleoprotein filament on the ssDNA and catalyzes the invasion of the ssDNA into a sister chromatid or homologous chromosome resulting in the formation of a joint molecule that acts as a primer for DNA synthesis to repair the damage

[77,78]. Finally, Holliday junctions formed during strand invasion are resolved, with or without sequence exchange [79], and rejoined by ligase XRCC4 (Fig.3) [78]. 17

Figure 3: The Major DSB DNA Repair Pathways: NHEJ & HR

NHEJ & HR are two pathways through which DNA DSBs are repaired. (A) NHEJ is favored by cells in G0/G1, and involves the direct re-ligation of the broken DNA ends. The initial event in NHEJ involves the binding of the Ku70/80- heterodimers to the broken DNA ends, following which the Ku-heterodimers come together forming a ring shaped structure to bridge matching DNA ends and in doing so recruit DNA-PKcs. DNA-PKcs then phosphorylates target proteins with endonuclease activity required to produce compatible DNA ends. Ends are then relegated by the XRCC4/DNA Ligase IV complex. (B) HR predominates over the other DSB repair pathways during the S and G2 phases of the cell cycle in which sister chromatids are readily available. HR is initiated by nucleolytic processing of the DSB to generate single-stranded DNA (ssDNA) overhangs which are then coated by RPA. Rad52 is then targeted to the newly processed ends by interaction with RPA, Rad52 in turn recruits the recombinase enzyme Rad51. Rad51 is targeted to sites of DNA damage through interaction with BRCA2, once released from BRCA2, Rad51 forms a nucleoprotein filament on the newly processed ssDNA overhangs, and then searches within the context of the nucleoprotein filament for an intact sister chromatid which acts as a template for DNA synthesis to repair the DSB. Holliday junctions formed during strand invasion are then resolved and rejoined by the ligase XRCC4.

1.5.3 Single-Strand Annealing (SSA):

A DSB repair pathway considered transitional between NHEJ and HR is SSA. SSA, like

HR, involves the use of homologous sequences for DSB repair, and like NHEJ is a Rad51- independent process that involves the annealing of DNA strands following resection at the DSB

[6]. The major players involved in the process remain somewhat unclear, however many members of the HR pathway, Rad52 in particular appear to be active in the process [74]. Following the induction of DNA damage the ends of the DNA DSB are processed by an exonuclease, research suggests the MRN complex [71], yielding long ssDNA overhangs. Once homologous ssDNA regions are exposed in the overhangs, they are annealed, the protruding ends are trimmed by the

ERCC1/XPF nuclease, and the gap is filled by DNA polymerase [74]. Like NHEJ this process is error-prone as it results in the deletion of unique intervening DNA sequences.

1.6 BRCA1 & BRCA2 Function in the DNA Damage Response

The ability to precisely control the order and timing of cell cycle events is essential for maintaining genomic integrity and preventing mutations able to disrupt normal growth controls.

Cells exposed to DNA damaging agents, such as ionizing radiation, coordinately arrest the progression of the cell cycle at the G1/S phase, the S phase and the G2/M phase to allow adequate time for damage repair [80,81]. It is now widely accepted that both BRCA1 and BRCA2 play multiple critical roles in the maintenance of genome stability as evidenced by a profound number of chromosomal translocations, duplications, and aberrant fusion events between non-homologous chromosomes in BRCA1 and BRCA2 deficient cells [18,82,83 ].

1.6.1 BRCA1 Function in DNA Damage Repair

Recent studies reveal that BRCA1 plays essential roles in HR repair, NHEJ, and nucleotide excision repair (NER) [84]; BRCA1 mediates these functions through interaction with components of the DNA repair machinery, and by regulating the expression of genes involved in the DNA damage repair pathways [84]. Studies suggest that BRCA1 carries out these roles in damage repair by acting as a link between the sensing and effecting components of the DNA damage response [6,82].

BRCA1 plays a critical role in responding to DSBs through its function in HR. In ways still to be defined, BRCA1 recruits BRCA2, which facilitates Rad51 filament formation on the ssDNA [86]. Rad51 catalyzes the invasion of the homologous sequence on the sister chromatid, which is then used as template for accurate repair of the broken DNA ends [87,88]. Other studies have shown that BRCA1 co-localizes with Rad50, a member of the MRN complex, following the induction of DNA damage; Mre11 encodes nuclease activity which resects flush ends of DSBs to generate ssDNA tracts [89]. BRCA1 binds DNA directly and inhibits this Mre11 activity regulating the length and the persistence of ssDNA generation at sites of DNA damage [90]. As ssDNA is a substrate for DNA repair by HR, it appears that BRCA1 might play an essential role in HR-mediated repair of DSBs through its inactivation of Mre11; an idea confirmed by the observation the HR is defective in BRCA1-deficient cells [85].

In addition to its somewhat unclear roles in DSB repair, BRCA1 has also been found to be a constituent member of a large nuclear protein complex named the BRCA1-associated surveillance complex (BASC). This complex contains the MRN complex, DNA-mismatch repair proteins MSH2, MLH1 and MSH6, DNA helicase BLM, ATM, RFC and PCNA [42]. Many of these proteins are involved in the sensing and repair of abnormal DNA structures, and have been

linked with the repair of replication-associated DNA damage [42]. Taken together this evidence suggests that BRCA1 might function as a coordinator of multiple processes required for the maintenance of genome integrity during the process of DNA replication and DNA-replication associated repair [24].

1.6.2 BRCA2 Function in DNA Damage Repair

BRCA1 and BRCA2 co-localize with Rad51 to form DNA repair complexes [91]. This co-localization of the BRCAs with Rad51 at sites of recombination and DNA damaged-induced foci strongly suggests that the BRCA proteins play a role in the detection of DSBs [91]. The roles played by BRCA1 and BRCA2 in the repair of DSBs by HR appear to differ, as evidence indicates a more direct role for BRCA2. The physical interaction between BRCA2 and Rad51 is essential for HR repair of DSBs to take place; BRCA2 is thought to be required for the transport of Rad51 from its site of synthesis to the site of DNA damage, where Rad51 is then released to form the nucleoprotein filament required for HR to take place [86]. Their interaction is mediated by the BRC repeats, and an unrelated domain located at the C-terminus of the BRCA2 protein

[22,23]. In BRCA2 defective cells, a reduction in the accurate repair of DSBs by HR is observed, along with an increase in the number of deletion events [82]. These deletions are thought to arise predominantly due to the shunting of DSB repair down the SSA repair pathway [82]. These cells also show an elevated sensitivity to IR, while the cell cycle checkpoint and apoptotic responses to

DNA damage remain intact [82,92]. Thus, while the function of BRCA2 in HR is well characterized it remains to be seen whether BRCA2 participates directly in cell cycle regulation

via checkpoint function, or whether it simply plays an indirect role in activating cell cycle checkpoints through a deficiency in the repair of DNA lesions [93].

1.7 Microarrays

Traditionally, hybridization methods attempt to identify and quantify only one gene or transcript at a time; these early membrane-based methods have more recently evolved into high through-put quantitative methods using fluorescence detection. Microarray technology allows large numbers of DNA clones with known sequences to be immobilized on a solid surface which then acts as an array of detection units (probes), while the pool of RNAs to be examined (targets) are fluorescently labeled and then hybridized to the probes [94]. Various platforms differing slightly in their construction are available for use; our research utilized the Agilent Inkjet -printed oligonucleotide chips the details of which are discussed below.

1.7.1 The Components of a Microarray

Printed oligonucleotide arrays are created by the robotic spotting and immobilization of synthetic oligonucleotides (oligos) onto a platform [95]. To create probes, genes of interest are chosen from public sequence databases including GeneBank, dbEST, and UniGene; several commercial oligonucleotide sets varying in probe length between 30 and 70 nucleotides are available for human samples [94]. Agilent arrays are unique as their probes are created using non- contact printing instead of robotic spotting, thereby reducing the variability of the features [94].

The probes are fired onto a solid platform using Inkjet technology in which small drops of

solution can be sprayed with high precision onto a surface [94]. Once the probe set has been selected, the targets are produced; this involves the labeling and amplification of sample RNA. In the simplest labeling strategy, a fluorescent dye-modified nucleotide is incorporated in the reverse transcription of mRNA to cDNA. The reverse transcription needs to be primed by a hybridized short oligonucleotide primer. The primer is usually an oligo-dT (12-18 nucleotides long) that anneals to the poly-A tail of eukaryotic mRNAs [94,95]. Fluorescent dyes most commonly used for labeling are the Cy3 (green) and Cy5 (red) dyes; in cases where the biological samples are so small that such RNA quantities cannot be extracted in vitro transcription-based methods for amplifying the RNA exist [94]. Finally, following the labeling and amplification of target RNA, hybridization of the labeled targets to the probes is performed by adding the targets dissolved in hybridization buffer to the slide, followed by a platform-specific incubation period.

Following hybridization an image of the array is acquired using a microarray scanner; these scanners have confocal lasers that produce light at the wavelength required to excite the fluorescent dyes, the fluorescence emission intensity of the dyes is then captured in high- resolution monochrome images [94]. For each spot on the array, the relative amount of fluorescence from each dye hybridized to its target is measured and the scanner software then displays a composite colored image [94].Finally, the fluorescence intensities are quantified from the high-resolution image. Because the two dyes have different properties and light sensitivities, the Cy5 dye is much more readily broken down resulting in a lower fluorescence emission [95], the fluorescence signals from the two dye channels have to be normalized in order to calculate correct expression ratios. Normalization corrects for different dye properties as well as for concentration differences between the co-hybridized test and reference samples.

1.7.2 Normalization

Following the acquisition of the fluorescence intensity images normalization procedures are carried out to ensure that differential expression observed between treatment conditions is due strictly to biological variation, and not biases introduced by the microarray platform [96].

Normalization methods must be selected with reference to the kind and degree of systematic bias present. Often a logarithmic transformation is used for microarray data as it tends to provide values that are approximately normally distributed (Gaussian) and therefore appropriate for conventional linear regression and hypothesis testing models [96].

1.7.3 Hypothesis Testing

Deciding whether a particular gene is differentially expressed across experimental conditions is often the end-goal of microarray experiments. Hypothesis testing can be employed to determine whether expression between conditions differs in a statistically significant manner.

An underlying assumption of hypothesis testing is that the data is roughly Gaussian in distribution; log transformation of data provides a relatively Gaussian distribution and hence is frequently employed prior to further analysis [94]. The classic test alternatives for a gene may be stated generically as: (a) the null hypothesis H0: gene (g) is not differentially expressed, and (b) the alternative hypothesis H1: gene (g) is differentially expressed [94]. A statistical test of hypotheses has two types of error: type I and type II. A type I error (false positive) occurs if the alternative hypothesis (H1) is concluded when, in fact, the null hypothesis (H0) is true. A type II error (false negative) occurs if (H0) is concluded when, in fact, the alternative (H1) is true [94]. A principal aim of microarray studies is to be able to declare a gene as differentially expressed with 25

high probability if it is truly differentially expressed, while keeping the probability of making a false declaration of differential expression acceptably low [95]. The p-value of the statistic is defined as the level of significance of the test for which H0 would just be rejected. Thus, the p- value for a gene (g) is a measure of how dissonant the evidence is with the null hypothesis, with smaller values being evidence against H0, in favor of H1[95].

1.8 Classification Methods

1.8.1 Unsupervised Learning

Statistical learning methods can be divided into two general classes, namely, supervised and unsupervised learning. In unsupervised learning, no predefined reference labels are used and classifications are made independent of prior knowledge [94]. Unsupervised methods are considered as useful approaches when there is little prior knowledge of the expected gene expression patterns for any condition. A commonly used method of unsupervised classification is that of hierarchical clustering in which data points are forced into a strict hierarchy of nested subsets [94]. Hierarchical clustering begins by considering each individual point as a cluster by itself. The procedure begins with (N) clusters (in the case of clustering genes) and successively combines the two closest clusters thereby reducing (N) by one in each step [95]. Results of hierarchical methods are shown in a tree diagram, known as a dendrogram. Several linkage methods are available for clustering, the most commonly employed of which is average linkage; average linkage clustering uses the (arithmetic) average distance measure for inter-cluster distances [95]. Hierarchical clustering is helpful for identifying relationships amongst gene sets or similarity amongst samples.

1.8.2 Supervised Learning

Conversely, supervised classification is a learning approach in which the classes of a subset of samples are inputs for the algorithm [94]. Essentially supervised classification involves the creation of a solution to a classification problem using the known classes of a subset of the samples comprising the dataset: the training set. Following ―training‖, the remaining samples, the test set, whose classes are presumed to be unknown by the algorithm, are predicted to belong to one class or another based on the similarity of their properties to samples in the training set. One challenge of classification using microarray data is that the number of features (genes) is significantly greater than the number of samples. In this situation, it is possible to find both random and biologically relevant correlation of gene behavior with sample type. To prevent against spurious results, the goal of a robust molecular classifier is to identify the smallest possible subset of genes that correlate most strongly with the known class labels. In addition, a small subset of genes is desirable for the development of expression-based diagnostic tools [97].

The Support Vector Machine (SVM) is a commonly used tool for the supervised classification of microarray data, and was the analysis tool used in this study. Briefly, based on the characteristics (gene expression profiles) of samples in the training set, a line or hyperplane

(non-linear case) that best separates the two classes of samples is found by maximizing a distance measure known as the margin (Fig.4) [98,99]. The margin describes the perpendicular distance from the hyperplane to the nearest sample of each class, known as Support Vectors (SV). Thus the optimal solution to the classification problem, the hyperplane, is the line that lies equidistant between the SVs of each class (Fig.4) [98,99]. SVM can be linear, or they can be representations of the data in a higher space; this is achieved by mapping the gene expression data to a higher-

dimensional space through use of a kernel: a mathematical function describing the transformation

[99].

1.8.3 Recursive Feature Removal

An inherent problem to microarray studies it that a typical gene expression data set is extremely sparse. That is to say, the gene expression data contains only dozens of samples but tens of thousands of features. This extreme sparseness is problematic as it is believed to reduce the performance of a classifier significantly [100]. As a result, the ability to extract a subset of informative features while removing irrelevant or redundant ones is critical for accurate classification [100]. Furthermore, identification of underlying carcinogenic molecular mechanisms is of key importance in developing more accurate diagnostic methods and therapeutic treatments. Recursive feature elimination algorithms are based on the theory that a relevant dimensionality reduction method should remove irrelevant/redundant features while keeping those that are most informative for classification purposes [100]. They are designed to effectively eliminate most of the irrelevant, redundant, and noisy genes present in the data set while keeping information loss to a minimum. Recursive feature elimination is dually advantageous in that a small feature set is highly desirable if one wishes to use the results of feature selection and classification to develop diagnostics tools, and secondly, the removal of redundant genes reduces the impact of over-fitting, and hence, potentially improves classification accuracy [97].

Figure 4: Pictorial Representation of an SVM

Based on the characteristics (gene expression profiles) of samples in the training set, a line or hyperplane (non-linear case) that best separates the two classes (class A & class B above) of samples is found by maximizing a distance measure known as the margin (m). The margin describes the perpendicular distance from the hyperplane to the nearest sample of each class, known as Support Vectors (SV). Thus the optimal solution to the classification problem, the hyperplane or decision boundary, is the line that lies equidistant between the SVs of each class.

1.9 Hypothesis & Objectives

It is known that one defective copy of BRCA1 or BRCA2 in the germ-line causes predisposition to developing early onset breast cancer [18]. As well, both BRCA proteins are known to play a role, either directly or indirectly, in DNA damage repair following exposure to IR and in the maintenance of genome integrity through participation in the cell cycle checkpoints.

From this evidence it is therefore reasonable to hypothesize that down regulation of these genes, through loss of a single copy, could have measurable effects in multiple critical pathways, and that these effects could be visualized as changes in the expression of a large group of genes. Thus, the overall goal of this work is to determine if individuals heterozygous for BRCA1 or BRCA2 germline mutations can be readily indentified from control individuals using gene expression profiling in both the baseline and their radiation-dependent cases.

These goals were carried out using three specific objectives. The first objective of this study was to elucidate the conditions for the assay, namely, an IR dosage and a recovery time at which gene expression changes related to induction of BRCA1/BRCA2 would be most apparent.

The second objective involved the treatment of all 97 lymphoblastoid cell lines (LCLs) used in this study to either the optimal dose of IR or mock-irradiation following which the cell lines were allowed to recover for the optimal recovery time. Following recovery, the total RNA of each sample was extracted, fluorescently labeled, and bound to Whole Human Genome Agilent DNA

Chips containing approximately 41,000 unique features. The third objective of this study was to examine the acquired expression profiles to determine if groups of genes with expression patterns that accurately distinguish BRCA1/BRCA2 mutation carriers from control individuals existed either in the baseline condition, following treatment with IR or in both.

Chapter 2: Materials & Methods

2.1 Cell Culture

EBV-transformed lymphocytes were obtained through the NIH Breast Cancer Family

Registries. The 97 cell lines included in this study consisted of 38 control (+/+), 31 BRCA1 mutation carriers (+/-), and 28 BRCA2 mutation carriers (+/-). The mutations used in this study were primarily frameshift and nonsense mutations, but also included some missense and splicing mutations; an exact list of the BRCA1/ BRCA2 mutations used in this study can be seen in Table

1. All LCLs used in this study were cultured in RPMI-1640 media (Sigma Aldrich, Oakville,

ON) supplemented with non-heat inactivated 15% fetal bovine serum (FBS) (Sigma Aldrich). All

2 cell culture was carried out in 25cm flasks (Corning, Nepean, ON) at 37°C in 5% CO2 atmosphere. Cells were split in a 2:1 ratio until the desired cell number of 650,000/ml was reached.

2.2 DNA Damage Induction

DNA damage was induced in each cell line through exposure to IR, delivered by a 137Cs

Victoreen Electrometer (Atomic Energy of Canada, Mississauga, ON) at a dose rate of

0.52Gy/min.

2.3 Cell Cycle Experiments

2.3.1 Determination of IR treatment dose and Recovery Time

Three test cell lines, (1 control, 1 BRCA1 carrier, and 1 BRCA2 carrier) were used to determine the optimal dose of IR. The three test cell lines were cultured as described above, and

then treated with varying doses (1, 2, or 4 Gy) of IR. The dose of radiation eventually selected for use was found to be the lowest dose of IR capable of producing consistent cell cycle checkpoint activation. Using the selected dose of 2 Gy IR, the optimal recovery point was then determined.

Each cell line was split into two: one fraction was then mock-irradiated, while the other was treated with IR. The cells were allowed to recover for a total time period of 24 hr, with aliquots being collected for staining and subsequent analysis at zero hours and every two hours post- treatment with IR.

2.3.2 G1/S-Phase Checkpoint Activation Visualization via Flow Cytometry

Cell proliferation labeling reagent (10 µl) (Amersham Biosciences, Baie d'Urfe, QC) was added 60 min prior to harvesting at each time point. At the selected time points, cells were washed once in 5 ml PBS and fixed in 5 ml 70% ethanol for 24 h at 4°C. Following fixation the cells were washed in 5 ml PBS, spun down at 5°C, 1000 RPM for 5 minutes and treated with

0.1mg/ml RNase A (Bioshop, Burlington, ON) at room temperature (RT) for 20 min, then washed once with 3 ml PBS and spun down at 5°C, 1000 RPM for 5 minutes. The cells were then incubated in DNA denaturation solution (0.5% Triton X-100; 4N HCl) at 37°C, for 30 min, spun down as above, and neutralized by the addition of 0.1M sodium borate, pH 8.5 for 5 minutes.

Cells were then washed once with 3 ml wash buffer (0.5% Tween-20 in PBS) and spun down as above. FITC-conjugated α-Bromodeoxyuridine (BrdU) antibody (20 µl) (BD Biosciences,

Oakville, ON) was added to the cell pellet for 60 min at RT, following which cells were washed in 3 ml PBS, spun down as above, and resuspended in PBS containing 80µg/ml propidium iodide

(PI) (CedarLane, Hornby, ON). Samples were then analyzed for checkpoint activation using

EPICS ALTRA flow cytometer (Beckman Coulter, Mississauga, ON). Cell cycle checkpoint

activation was visualized through use of dual staining with BrdU (BD Biosciences) and PI

(CedarLane), followed by analysis using an EPICS ALTRA flow cytometer (Beckman Coulter).

2.4 Gene-Expression Profiling

The RNA from each of the 97 cell lines, one fraction treated with 2 Gy IR and one mock- irradiated fraction, was extracted at the same time following a recovery period of 6 hr. RNA was extracted using TRIZOL® Reagent as per the product protocol (Invitrogen, Burlington,ON); the

RNA was then further processed using the RNeasy MinElute Cleanup Kit (Qiagen, Germany).

RNA quality was determined using an Agilent 2100 Bioanalyzer (Version B.02.02); only RNA samples with an RNA Integrity Number (RIN) of 7or greater were selected for further experimentation. Following confirmation of RNA quality, RNA was labeled and amplified using the Agilent Low RNA Input Linear Amplification kit. As per the product protocol, 250 ng of control and experimental total RNA were labeled with Cy3-CTP and Cy5-CTP (Perkin Elmer,

MA, USA) and 2 µl of Agilent Spike-In mix was added to monitor the amplification and labeling processes. Following amplification, RNA was quantified using the NanoDrop ND -1000

(NanoDrop Technologies, DE, USA) and the concentration of cRNA and the specific dye activity were calculated. Those samples with a specific dye activity greater than 8 pmol/µl were selected for hybridization to the arrays. Samples initially failing to meet either the desired RIN or specific dye activity were either re-extracted, or re-labeled for inclusion in the study. 825 ng of both the

Cy3 and Cy5-labeled cRNA were then hybridized to the Agilent Whole Human Genome long- oligo 4x44K GE arrays as per the product protocol. Image acquisition and analysis were done using an Agilent Microarray Scanner, Model G2565BA and Agilent Feature Extraction software v9.1 set to default settings.

2.5 Data Analysis

2.5.1 Filtering & Normalization

A starting number of 43,376 features were present on the Agilent whole human genome microarrays. Following scanning, quality filtering of the raw expression data was carried out.

Poorly expressed genes, defined here as having a raw expression level < 5, and potentially saturated genes (raw expression level > 40,000) were removed from each dataset prior to further analysis. In the BRCA1/BRCA2 baseline expression analysis 23,921 and 23,553 features remained following filtering, and in the radiation-dependent case 22,489 and 22,275 remained respectively.

Normalization of the datasets was then carried out. In the case of the non-radiation dependent

(baseline) gene expression data, normalization was carried out by standardizing the raw expression values of each row (gene) to a mean =0 and a standard deviation =1. For the radiation- dependent gene expression data, the log base 2 ratio of the Cy5/Cy3, where the non-irradiated sample was labeled with Cy3 and the irradiated sample was labeled with Cy5, was taken following which the raw expression values of each row were set to a mean=0 and a standard deviation=1. Following normalization, the number of salient features was further decreased through the use of hypothesis testing. On a per gene basis, the mean of the control samples was compared to the mean of the BRCA1 or BRCA2 carrier samples using a feature called ―ttest2‖ in

MATLAB®(The Mathworks Inc., Natick, MA, USA) the genes identified as differentially expressed at p- values of 0.5, 0.1, 0.05, and 0.01 (where applicable) were then recorded. The number of genes associated with the p-value yielding the classifier with the highest initial accuracy rate was then selected as the starting point for further classification in each case. The initial modeling of the data for each of the four classifiers used all features remaining following the quality filtering process described above. For the baseline expression classifiers, this left

starting numbers of 41 and 94 features for BRCA1/ BRCA2 respectively. For the radiation- dependent classifiers the starting number of features was 1110 BRCA1 and 476 for BRCA2.

2.5.2 Classification Algorithms

Following quality filtering and normalization, the filtered gene expression data was analyzed using a radial-based support vector machine (SVM) coupled with a novel recursive feature removal algorithm implemented in MATLAB®. Implementation of the SVM was done using fixed training and test sets comprised of 51 training (22 wt, 29 BRCA1 (+/-))and 18 test (9 wt, 9 BRCA1 (+/-)) in the BRCA1 case, and 49 training (20 wt, 29 BRCA2 (+/-)) and 17 test (9, 8) samples in the BRCA2 case. The radial-based kernel used was of the form [exp(-Gamma*|x(:,i)- x(:,j)|2)]; code for the SVM algorithms used was written by Dr. Saeed Hashemi [101].

The effect of feature selection on prediction accuracy was then examined by recursively removing features found to be the least-discriminating between classes. To do so, features were ranked using a novel algorithm and then removed successively. The novel Recursive Feature

Removal (RFR) algorithm ranked each feature using the relative gene weights of the support vectors [SVs], and then removed features in an incremental fashion based on the theory that truly salient features would be selected more often than redundant or non-salient ones [100].

Specifically, all features remaining following classification were used to run an initial trial of the radial-based SVM. The output of this initial modeling, specifically the information contained in the file labeled (SVs) (support vectors) was then collected. In this output file the contribution of each gene to each support vector (SV) is recorded; this file is separated into two, the samples that act as SVs for one class and the samples that act as SVs for the opposing class. SVs are samples

that define the margins of the hyperplane: i.e. samples whose exclusion would change the given solution to the classification problem. The absolute value of the weights of each gene for each set of SVs were then summed and multiplied, yielding a single value for gene. The genes were then assigned an ordinal rank based on this single value, with 1 being the most-discriminate gene, and the largest ranking denoting the least-discriminate gene. Using the newly assigned ranks, the genes are then removed x% at a time (where x changes in each of 9 trials starting at 90%, decreasing by 10% per trial ending at a final removal rate of 10%). For each of the 9 trials, the instance, number, and identity of the genes yielding the highest accuracy rate classifier are recorded. Therefore, at the end of the 9 trials, the 9 best accuracy rates (one per trial) and the genes yielding these instances are recorded and totaled in a functional called ―tally‖. The next phase of the algorithm removes genes using the information stored in the ―tally‖ variable. Thus, the ―tally‖ function is a file containing genes denoted by numbers (1: n) where n= the number of genes remaining following filtering, and integers running from (0:9) denoting the number of times each gene was called. The post-tally algorithm then removes genes in 9 steps staring with those called once, and ending with those called nine times. After each removal the accuracy rate of the classifier is recorded. From here the list of genes yielding the highest classification accuracy is selected for further analysis by the algorithm. Here the genes are again ranked using the output

[SVs] as described above and removed one at a time. This step-wise single gene removal was found to remove genes initially included in the classifier that proved to be redundant or non- essential, as their removal improved the accuracy of the classifier. The algorithm finished by recording the lowest number of genes yielding the highest possible classification accuracy; a working example of the RFR algorithm described above can be seen in Figure 5.

Figure 5: Working Example of RFR Algorithm

The first step in the RFR process is a preliminary run of the SVM using all features included in the initial p-value threshold selected. In this example, the initial number of features is 10. Following the initial run of the SVM the output [SVs] each contain a weight corresponding to each of the ten genes. This output is then used to rank the importance of each gene in the classification task. Once the rank of each gene is calculated, the genes are reordered in a new table with 1 being the highest rank and n), in the example n=10, being the lowest. Nine trials removing 10-90% of the genes per trial (x=1, remove 90%, x=2, remove 80%...) are then run to create the ―tally‖ function. Follow the boxes in the working example from top to bottom. Process described in detail in the text.

2.5.3 Visual Representation of Data

Following the determination of the best-discriminating features for each of the classifiers heat maps were constructed using the Genesis software package [102].

Chapter 3: Results

3.1 Cell Cycle Checkpoint Activation is Present in all Genotypes

In order to maximize our ability to visualize any discrete changes related to the

BRCA1/BRCA2 mutation carrier genotype at the gene expression profiling level, the known roles of BRCA1/BRCA2 in the DNA damage response were exploited. Time point experiments examining cell cycle checkpoint induction were performed in order to elucidate a treatment dose of IR and a recovery time at which changes in BRCA 1/BRCA2 and their interacting partners’/downstream effectors expression would be most visible. Cell cycle checkpoints block the progression of a damaged template through the cell cycle. At key transitions during eukaryotic cell cycle, checkpoint signaling pathways monitor the successful completion of upstream events prior to proceeding to the next phase [38].

3.1.1 Determination of optimal IR treatment dose

Based on previous research conducted Drs. Davey and Feilotter [unpublished] and by

Foray et al. [103], doses of 1, 2 and 4 Gy of IR were initially selected for use. We found that cell cycle checkpoint activation occurred in test cell lines of all three genotypes using doses of both 2 and 4 Gy IR therefore the lower dose (2Gy) was selected in an effort to minimize cytotoxic effects.

3.1.2 Elucidation of Optimal DNA Damage Recovery Period

Following determination of the optimal dose of IR, time point experiments were conducted in order to identify a time point following exposure to IR that would allow the greatest possibility of witnessing BRCA1/BRCA2-related changes at the gene expression profiling level.

Cell cycle checkpoint activation was visualized through use of dual staining with BrdU and PI.

BrdU is incorporated in place of thymidine allowing visualization of newly synthesizing DNA, while PI, an intercalating agent, allows visualization of DNA content. Following treatment with 2

Gy IR all three test cell lines: 1 control, 1 BRCA1 mutation carrier and 1 BRCA2 mutation carrier were all found to have successful checkpoint activation, shown in Figure 6 as a loss of cells transitioning from G1 to S-phase, a decrease in the S-phase population and a corresponding increase in the G2/M-phase population. As shown in Figure 7, regardless of genotype, all cell lines consistently showed checkpoint activation at the 6 hour time point and a considerable amount of variability in their release from checkpoint arrest thereafter. This variance showed little correlation to genotype, as different cell lines of the same genotype were found to show individual differences in checkpoint release. The 6 h time point was found to be the longest recovery period following treatment with IR at which the cell cycle was consistently arrested in all cell types, and thus was chosen as the optimal recovery period for our study.

3.2 The DNA Damage Response is present in control cell lines

To ensure that the selected dose of 2 Gy IR was capable of activating a DNA damage response consistent with that seen in the literature, the expression profiles of the control samples prior to and following treatment with IR were examined using Significance Analysis of

Microarrays (SAM) software [104]. As seen in the abbreviated list shown in Table 1 (complete list with p-values found in appendix; S-T1) gene changes consistent with the literature pertaining to the DNA damage response following treatment with IR [104, 105, 106] were found.

Figure 6: Flow Cytometry Density-Plot Histogram

Visualization of cell cycle checkpoint induction via dual staining with BrdU-FITC (Y-axis) which incorporates itself in place of thymidine in newly synthesized DNA and PI (X-axis) a measure of DNA content. The bottom left box of each histogram labeled G1- contains the G1-phase cells population, the top box (S) represents the S-phase cell population, and the bottom right-hand box (G2M) represents the G2/M cell population. (A) Density plot histogram of mock-irradiated wild-type test cell line. (B) Density plot histogram of wild type cell line 6 h post-treatment with 2 Gy IR. Activation of G1/S cell cycle checkpoint is evidenced by the lack of cells transitioning from the G1 to S-phase, the loss of cells in early S-phase, and the corresponding increase in cells in late S-phase and the G2/M phase.

0h 4h 6h 8h 12h 18h

Figure 7: Flow Cytometry Density-Plot Histograms Visualization of cell cycle checkpoint induction in a three test cell lines, 1 wild type,1 BRCA1 (+/-) and 1 BRCA2(+/-). Visualization is via dual staining with BrdU-FITC (Y-axis) and PI (X-axis). Recovery time following treatment with IR is indicated below each column. At the 6 h time point cell cycle checkpoint activation is clearly visible as evidenced by a loss of cells transitioning from the G1 to S-phase, a decrease in the S-phase population and a corresponding increase in the G2/M phase population. Release from checkpoint arrest, seen as leakage of cells from G1 into S-phase, is seen to vary amongst the cell lines, with no correlation to genotype.

3.3 BRCA1/BRCA2 Mutation Carriers Are Readily Identifiable At The Gene-Expression Profiling Level Following The Induction Of DNA Damage By Treatment With IR

Previous studies investigating the use of gene expression profiling to distinguish carriers of functional BRCA1/BRCA2 mutations from control individuals suggest that the task is achievable with a high level of accuracy following the induction of DNA damage through exposure to IR [107,108]. Accordingly, our study sought to determine whether carriers of all classes of BRCA1/BRCA2 mutations could be readily identified using only gene expression profiling. To do so, the gene expression profiles of 38 control LCLs were compared with either 31

BRCA1 or 28 BRCA2 mutation carrier cell lines. Unlike previous studies investigating the potential of gene expression profiling for accurate identification of BRCA1/BRCA2 heterozygotes

[107,108] this study included cell lines harboring missense mutations classified as functional on the basis of data in the BIC [12]. All cell lines were exposed to 2 Gy IR, in order to induce DNA damage, following which the total RNA of all samples both without treatment and 6 hr post- treatment with IR was extracted, amplified, fluorescently-labeled, and affixed to Agilent long- oligonucleotide 4x44 K microarrays representing the entire human genome. The scanned microarrays were then analyzed and molecular classifiers comparing the BRCA1 mutation carrier expression profiles to the control profiles, and a second comparing the BRCA2 carrier profiles to the control profiles were constructed using a radial-based SVM in combination with a novel RFR algorithm. A flow chart explaining the construction of the classifiers is shown in Figure 8.

This study employed dual analysis methods; the first involved the construction of molecular classifiers using solely the gene expression data contained in the training set such that the samples in the test set could act as independent validation of our results. The second analysis method selected differentially expressed features using the entire set; results of this analysis are presented, yet require validating. 46

Figure 8: Flow Chart: Step-Wise Construction of Molecular Classifiers

The step-wise construction of the molecular classifiers is shown here. Expression data for the initial 43,376 features contained on the Agilent Whole Human Genome Chips were processed firstly by applying a quality filtering process. Genes with values outside the selected ceiling (saturation) and threshold (no expression) were removed. The features remaining in each of the four cases were then normalized by setting the mean expression of each gene = 0, and the S.D. = 1. Following normalization, hypothesis testing was carried out to identify the most salient features. The datasets were then split ~2/3:1/3 training to test sets and the test sets were used to train the initial SVM. From here, features were removed in a multi-step process using a novel Recursive Feature Removal (RFR) algorithm.

3.3.1 Analysis I: Construction of BRCA1/BRCA2-Radiation Dependent Molecular Classifiers with Independent Validation

Firstly, the selection of differentially expressed genes using data contained in the training set only was undertaken. As shown in Table 3, as the stringency of the p-value threshold was increased, the accuracy of the classifiers tended to decrease; thus a p-value≤0.05 was selected as a starting point for the BRCA1/BRCA2 IR-dependent classifiers. At this p-value threshold, 1110 and

476 salient features for BRCA1/ BRCA2 respectively remained; a list of these genes and their associated p-values is found in the appendix (S-T2, S-T3). Notably, these gene lists were found to contain a number of known tumour suppressors and oncogenes, as well as genes known to play a role in cell cycle regulation and the DNA damage response pathways. Of interest in the BRCA1 list were: RAD23B, RAD50, RAD17, RB1, CDKN1B (p27) and STAT1 while the BRCA2 list contained SWI/SNF related SMARCE1, RB1, FGF2, and BRCA2. Four of these genes have previously been shown to be differentially expressed between BRCA1/BRCA2 mutation carriers and wild type individuals following exposure to IR [107,108].

The SVM algorithm considers all selected genes to create descriptions of samples in high- dimensional space and then defines a hyperplane that best separates the samples from two disparate classes [109] thus creating a classifier able to distinguish BRCA1/BRCA2 mutation carriers from control individuals.

The initial SVM-modeling of the IR-dependent expression data used all genes differentially expressed between the BRCA1 or BRCA2 mutation carriers and the control samples at the selected p-value threshold. The BRCA1-IR classifier trained on the initial 1110 features yielded a preliminary accuracy rate of 44%, while the BRCA2-IR classifier trained on the initial

476 features yielded an accuracy rate of 59% (Table 3).

The accuracy of the IR-dependent classifiers was improved by coupling the radial-based

SVM to a novel RFR algorithm that used the output of each successive run of the SVM to remove the least-discriminative features (described in detail in Materials & Methods [2.5.2]).

Implementation of the RFR algorithm was able to reduce the number of salient features in the

BRCA1-IR classifier from 1110 to 12 yielding a final accuracy rate of 67%, while the BRCA2-IR classifier was reduced from 476 to 37 features yielding a final accuracy rate of 77% (Table3).

Following the point at which the lowest effective gene number was reached, the accuracy of the classifiers was seen to fall off dramatically. The final BRCA1-IR classifier of 12 genes was found to classify 12/18 (67%) test samples correctly, with errors comprised of 3 false negatives (FN) and 3 false positives (FP), whiles the final BRCA2-IR classifier of 37 genes was found to classify

13/17 (77%) test samples correctly, with errors comprised of 4 FN (Table 3).

3.3.2 Analysis II: Construction of Radiation-Dependent Molecular Classifiers Using the Entire Dataset for Feature Selection

The classifiers discussed above were created using features identified as differentially expressed between BRCA1/BRCA2 mutation carriers and control individuals based solely on the expression data contained in the training set. Here instead, the entire data set was considered in the selection of salient features, resulting in an improvement in the classification ability of the IR- dependent classifiers, but lacking validation. Hence forth classifiers trained on the entire dataset will be denoted by the inclusion of the letters ―AD‖ (All Data) in their title.

As shown in Table 3, as the stringency of the p-value threshold was increased, the accuracy of the IR-dependent classifiers tended to decrease. As such, a p-value threshold of 0.05 was selected as a starting point for further analysis yielding starting numbers of 907 and 502

salient features for BRCA1 and BRCA2 respectively (S-T4, S-T5). As above, these gene lists were found to contain a number of known tumour suppressors and oncogenes, as well as genes known to play a role in cell cycle regulation and the DNA damage response pathways. The BRCA1 list was found to contain: BARD1, BRCA1, CAV1, CDKN1B (p27), CREBBP, LEF1, RAD23B,

RAD50 and RB1. While the BRCA2 list was found to contain: FGF2, RAD50, SMARCE1 and

SYK.

The BRCA1-IR-AD classifier trained on all 907 features yielded an initial accuracy rate of

72%, while the BRCA2-IR-AD classifier trained on 502 features yielded an initial accuracy rate of

77% (Table3). As above, classification accuracy was further improved by implementation of the

RFR algorithm. Implementation of the RFR algorithm reduced the number of salient features in the BRCA1-IR-AD classifier from 907 to 94, correctly classifying 16/18 (89%) test samples, with errors comprising 1 FN and 1 FP, while the BRCA2-IR-AD classifier reduced the initial 502 salient features to 54, correctly classifying 16/17 (94%) test samples yielding only 1 FN (Table 2).

3.3.3 Comparison of Features Selected in Radiation-Dependent Analyses I & II

A comparison of the features comprising the final IR-dependent BRCA1/BRCA2 classifiers in the first and second analyses reveals a high degree of overlap in gene membership.

As shown in Table 4, seven of the 12 features comprising the final BRCA1-IR classifier were also found in the BRCA1-IR-AD classifier, while fifteen of the 37 features comprising the BRCA2-IR classifier were also present in the BRCA2-IR-AD classifier.

TABLE 4: Comparison Of Genes Comprising Final IR-Dependent Classifiers Based on Analyses I&II BRCA1 vs WT-IR BRCA1 vs WT-IR BRCA1 vs WT-IR BRCA2 vs WT-IR BRCA2 vs WT-IR Final Classifier Final Classifier Final Classifier Final Classifier Final Classifier 94 Features 94 Features 12 Features 54 Features 37 Features Accuracy=89% Accuracy=89% Accuracy=67% Accuracy=94% Accuracy=77% ALL DATA ALL DATA TRAINING SET ONLY ALL DATA TRAINING SET ONLY A_24_P110601 HEAB APOL6 A_24_P307046 A_24_P565898 A_24_P50509 KIAA0738 C9orf3 A_24_P931711 A_24_P931711 A_24_P635355 LNX2 CCDC55 AATK ADAM33 A_24_P75688 LOC133874 FAM98A ACBD3 AK057740 AGER LOC153346 HEAB ADAM33 AW235815 AK057591 LOC285535 HNRPD AHCTF1 B4GALNT1 AK123765 LOC51136 MGC52282 B4GALNT1 C16orf63 APOL6 MALT1 PTD004 C11orf68 C7orf28A ARIH2 MAP3K10 RASSF2 C8orf38 CCNT2 BC000206 MAPK11 SAPS2 CCDC117 CDCA7 BC038355 MGC16121 SIPA1L1 CEP290 CHMP4B BE044472 MGC33556 TANC1 CHMP4B CRBN BRCA1 N75427 CUL4A ELF1 BRCA1 N75427 DKFZP586P0123 ELF4 BRDG1 NAT12 DLST ENST00000297544 BXDC2 NF529 DNMT3A ENST00000314295 BXDC2 NOC3L ELF1 FBXO34 C10orf86 P73L ELF4 GOLPH3L C10orf86 PCYOX1 EMD KRT9 C9orf3 PHF5A ENST00000292562 LOC222171 CCDC55 PICALM ENST00000297544 LOC285033 CCL5 PITPNB ENST00000327423 MYO1A CDC37L1 POLR2J ENST00000376573 NSMCE2 CDV3 PRR4 FBXO34 NY-SAR-48 CF528315 RAB40B FGF2 PARN COL5A2 RAC1 FLJ12986 PIN4 CREBBP RAC1 GOLPH3L RAB6IP2 CREBBP RAC1 K095727' RCC2 CREBBP RAC1 KAZALD1 RP11-308D16.4 CREBBP RAC1 KIAA0082 SFXN2 CREBBP RASSF2 LOC285033 SLC39A8 CREBBP RHEB MGC70870 SPIN CREBBP RMND5A MSN TMEM85 CRELD1 SCAMP5 MYO1A TNFSF5IP1 CTCF SLA/LP NPR2 TXNDC12 DB379047 ST7L PDCD6IP WRB DDX27 TANC1 PRPH XKR8 ELF4 THC2279548 PYGO2 ENST00000259969 TM7SF3 RAB6IP2 ENST00000297544 TNFSF5IP1 RIOK3 ENST00000345365 TRERF1 RKHD2 ENST00000370238 TRIM22 RSL1D1 ESR2 UBE2H SECISBP2 FAM98A UROD SLC39A8 FANCF USP45 SYK FLJ21908 ZAP70 TBPL1 GPC2 ZFP64 THC2313495 *Features highlighted were selected as final features in Analyses I&II TNFSF5IP1 * Certain features contained more then one probe set on the TTC28 microarrays and thus can be called more than once TXNDC12 UBE2Q1 VPS45A WIPI2 ZDHHC11

3.4 BRCA1/BRCA2 Mutation Carriers are readily Identifiable through Gene Expression Profiling at the Baseline Level

The proteins encoded by the BRCA1 and BRCA2 genes are known to function in the repair of DNA damage occurring both endogenously, during DNA replication, and exogenously as a result of exposure to genotoxic agents; we hypothesized that changes related to endogenous repair deficiencies would be visible in the baseline gene expression profiles of BRCA1/BRCA2 mutation carriers. Whereas previous studies examining the utility of baseline expression profiling to distinguish carriers of BRCA1/BRCA2 mutations from wild type individuals reported no significant differences in the baseline gene expression profiles of these two groups [107,108] we found that the two groups are readily distinguishable using baseline gene expression profiling.

3.4.1 Analysis I: Construction of BRCA1/BRCA2 Non-Radiation Dependent Molecular Classifiers with Independent Validation

The gene expression profiles of 38 control, 31 BRCA1 and 28 BRCA2 mutation carrier

LCLs were analyzed in order to determine if subtle changes in gene expression associated with the BRCA1/BRCA2 carrier genotype existed at the baseline level and were capable of distinguishing mutation carriers from control individuals. As described above, molecular classifiers comparing BRCA1/BRCA2 mutation carriers with control samples were constructed.

Unlike the radiation-dependent expression data, the accuracy of the baseline BRCA1/BRCA2 classifiers tended to improve as the p-value threshold became more stringent; thus starting thresholds of p≤0.001 and p≤0.005 for BRCA1 and BRCA2 respectively were selected for the initial SVM-modeling of the data. Selection of these thresholds yielded starting numbers of 41 and 94 features for BRCA1/BRCA2 respectively. 54

The BRCA1 classifier constructed using these 41 features yielded an initial classification accuracy rate of 78%, while the BRCA2 classifier achieved an initial accuracy rate of 71% (Table

5). Again, through implementation of the RFR algorithm, the number of salient features was reduced while classification accuracy was increased. Specifically, in the case of BRCA1, implementation of the RFR algorithm reduced the initial 41 features to 33 capable of correctly classifying 15/18 (83%) test samples, with errors comprising 1 FP and 2 FN; the BRCA2 classifier was reduced from 94 to 23 features correctly classifying 13/17 (77%) test samples, with errors comprising 2 FP and 2 FN (Table 5).

The preliminary BRCA1/BRCA2 baseline expression classifiers comprised of 94 and 41 features respectively were then further examined through use of hierarchical clustering in Genesis

[102]. As shown in Figures 9 &10, differences in the expression of these top-discriminating features are clearly visible between both BRCA1 and BRCA2 mutation carriers and their wild type counterparts.

Finally, as shown in Figure 11, the features comprising the initial BRCA1/BRCA2 baseline classifiers were sorted into their gene ontology categories in order to establish their functional significance [110]. Interestingly, both classifiers were found to contain a large number of genes involved in cell communication, proliferation and differentiation. Also of note was the large proportion of genes (14% & 11%) in both classifiers involved in biological regulation of key cell processes including transcription, cell proliferation, and apoptosis.

Figure 9: Hierarchical Clustering of Features Comprising BRCA1 Non-IR Classifier Hierarchical clustering of the 41 features selected as most differentially expressed at a p≤0.001. Selection of differential genes was done using gene expression data contained in the training set only. Legend: 0: WT Sample, 1: BRCA1 (+/-) Sample

Figure 10: Hierarchical Clustering of Features Comprising BRCA2 Non-IR Classifier Hierarchical clustering (average linkage) of the 94 features selected as most differentially expressed at a p≤0.005. Selection of differentially expressed features was done using gene expression data of the training set only. Legend: 0: WT Sample, 2: BRCA2 (+/-) Sample.

Figure 11: Gene Ontology Categories of Features Comprising the BRCA1/BRCA2 initial Non-IR Classifier

A) The 41 features comprising the initial BRCA1 (non-IR) classifier, identified at a p-value threshold of 0.001 or less were grouped into their gene ontology categories. Note the large percentage of genes with functions in cell proliferation, differentiation, and communication as well as transcription. B) The 94 features comprising the initial BRCA2 (non-IR) classifiers, identified at a p-value threshold of 0.05 were grouped into their gene ontology categories. Note the similarities in function of these genes compared to those comprising the BRCA1classifier.

3.4.2 Analysis II: Construction of BRCA1/BRCA2 Non-Radiation Dependent (Baseline) Molecular Classifiers

As in the case of the IR-dependent classification described above, a second analysis using the entire data set for the selection of salient features was undertaken. Here again the selection of differentially expressed features using the entire dataset was found to improve the accuracy of the baseline expression classifiers. Unlike in the first analysis of the baseline gene expression data, here the imposition of more stringent p-values did not show a clear trend on accuracy rate (Table

5); thus the p-value yielding the highest classification accuracy rate using the fewest number of features was selected as the starting point for data modeling. In both cases this led to the selection of p≤0.05, yielding starting numbers of 144 and 153 features for BRCA1/BRCA2 respectively. The initial baseline BRCA1-AD classifier correctly classified 16/18 (89%) test samples, with errors comprising 2 FN, while the initial BRCA2-AD classifier correctly classified 12/17 (71%) test samples with errors comprising 2 FN and 3 FP. Surprisingly, implementation of the RFR algorithm in the BRCA1-AD baseline classifier was found to significantly reduce the number of salient features from 144 to 25 without improving the classification accuracy. In contrast, implementation of the RFR algorithm was found to significantly increase the classification accuracy of the baseline BRCA2-AD classifier from 71% to 82%, correctly classifying an additional two test samples (Table 5).

3.4.3 Comparison of Features Selected in Non Radiation-Dependent Analyses I & II:

Upon examination of the features comprising the BRCA1 non-IR classifiers from the first and second analyses it can be seen that all features comprising the BRCA1 non IR-dependent

classifier created using solely the training set expression data are also contained in the BRCA1 non-IR-AD classifier (Table 6). While as shown in Table 6, of the 94 features comprising the training set only-derived BRCA2 non-IR classifier 74 were found to be present in the BRCA2 non-

IR-AD classifier.

3.5 Feature Selection for Classification versus Biological Insight

As the molecular classifiers discussed above were designed to select the smallest number of features capable of attaining the highest possible classification accuracy it is likely that the genes comprising these final classifiers are not providing a total picture of the underlying molecular mechanisms at work. Thus, in order to gain some insight into the effects of

BRCA1/BRCA2 heterozygosity following exposure to IR a third analysis was conducted.

Using SAM [104], the gene expression profiles of the 31 BRCA1 mutation carriers and the 28 BRCA2 mutation carriers were examined separately for radiation-induced changes in gene expression. Complete lists (S-T6, S-T7) of these genes and their associated p-values can be found in the appendix. These lists were found to contain largely the same genes as those seen in the post-IR wild type dataset. Interestingly however, in the case of BRCA1, BRCA1 and BARD1 were both found to be down-regulated. While in the case of BRCA2, genes involved in the HR repair pathway, namely RAD51 and BRCA2 were found to be down-regulated. Of these changes only the down-regulation of BARD1 and RAD51 was found in the wild type case, however to a much lesser degree (see S-T1, S-T7 & S-T8).

TABLE 6: Comparison of Non-IR Dependent Features Selected in the Training Set Only Classifiers & All Data Classifiers BRCA1 vs WT BRCA1 vs WT B1 vs WT BRCA1 vs WT BRCA2 vs WT BRCA2 vs WT BRCA2 vs WT BRCA2 vs WT BRCA2 vs WT 144 Features 144 Features 144 Features 41 Features, 153 Features 153 Features 153 Features 94 Features 94 Features Accuracy=89% Accuracy=89% Accuracy=89% Accuracy=78% Accuracy=71% Accuracy=71% Accuracy=71% Accuracy=71% Accuracy=71% Training Set ALL DATA ALL DATA ALL DATA Training Set Only ALL DATA ALL DATA ALL DATA Training Set only only A_23_P103951 ENPP2 MSX1 A_32_P26721 A_23_P103951 CTTN NFKBIZ A_24_P144314 LOC256021 A_24_P144314 ENST0000262525 MTAC2D1 AW327568 A_24_P144314 DLL1 NID1 ADARB2 MFSD7 A_32_P18630 ENST0000262525 MX2 BCR ADARB2 DOK4 NMT2 AMICA1 MGC22014 A_32_P26721 ENST0000285206 MX2 BCR AGPAT4 EBI3 NTN1 AW327568 MIA A_32_P3221 ENST0000292729 NFATC4 BCR AGPAT4 EFNA1 PACSIN3 AY358802 MLF1 ABTB2 ENST0000301807 NFKBIZ BCR AHNAK EMID1 PDE4B BRCA2 NFATC4 ADARB2 ENST0000301807 NID1 C21orf30 AIRE ENPP2 PELI2 BRCA2 NID1 AW327568 ENST0000321464 OASL CSRP2 AK023774 ENST0000227451 PGEA1 BRCA2 NTN1 BC082970 ENST0000371276 OGFRL1 EMID1 AY358802 ENST0000262525 PGF BRCA2 P2RX5 BCAR3 ENST0000379156 PARP12 EMID1 BC082970 ENST0000262525 PITPNC1 BRCA2 PDE4B BCL2L11 ENST0000382108 PC ENST0000262525 BE175081 ENST0000285206 PLA2G4A BRCA2 PELI2 BCL2L11 ETV7 PDCD1 ENST0000262525 BRCA2 ENST0000292729 PLA2G4A BRCA2 PGF BCR FAM79B PDE4B ENST0000285206 BRCA2 ENST0000325863 PLA2G4A BTBD3 PLA2G4A BCR FCGRT PDE4DIP ENST0000292729 BRCA2 ENST0000336749 PLA2G4A BTBD4 PLA2G4A BCR FLJ10357 PIM1 ENST0000301807 BRCA2 ENST0000344523 PLA2G4A C10orf54 PLA2G4A BCR FLJ31033 PLA2G4A ENST0000321464 BRCA2 ENST0000357776 PLA2G4A C14orf58 PLA2G4A BCR FLJ42953 PTK7 ENST0000337025 BRCA2 ENST0000375672 PLA2G4A C14orf79 PLA2G4A BCR FLJ42953 PTPRK ENST0000371276 BRCA2 ENST0000379748 PLA2G4A CA431756 PLA2G4A BG547557 FOXP1 PWWP2 FCGRT BTBD3 ENST0000382108 PLA2G4A CDC42EP2 PLA2G4A BHLHB5 FOXP1 RGS12 FLJ10357 BTBD4 FAM69B PLA2G4A CDC42EP4 PLA2G4A BI910665 GLDC SCD5 FLJ42953 C14orf58 FCGRT PRRT3 CETP PLA2G4A BTBD3 GMPR SDC4 FLJ42953 C14orf79 FKBP1B PTGIR CKB PLA2G4A BU685299 GPR155 SNN FOXP1 C1orf102 FLJ10357 PTK7 CLMN PTGIR C10orf54 HERC5 SOX4 FOXP1 C1orf162 FLJ36166 RASSF4 CNR1 RASSF4 C11orf75 HERC6 TBC1D16 GPR155 C20orf112 FLJ42342 RASSF4 CR596491 RASSF4 C1orf162 HERC6 TBX21 HERC6 C21orf30 FRY RGS12 DOK4 RHOV C1orf162 HEY1 THC2269172 IFI6 C6orf59 GADD45G ROBO1 DOK4 SDC4 C1orf53 ID3 THC2312637 IFIT1 CA431756 GRAMD1C SDC4 EMID1 SOX9 C1orf54 IFI6 THC2314822 IFIT2 CCDC19 GSN SOCS1 EMID1 SSTR2 C21orf30 IFIT1 THC2314823 IFIT3 CD274 HSD11B1L SOX4 EMILIN1 SSTR2 C9orf122 IFIT2 THC2406050 IFNA4 CD40 HSD11B1L SOX4 ENPP2 TA-NFKBH CCR10 IFIT3 THC2411070 LFNG CD40 IFIT3 SOX9 ENST0000227451 TBX21 CD24 IFIT5 THC2450504 LOC256021 CD40 IL18BP SSTR2 ENST0000285206 THC2266474 CD248 IFITM4P TMEM61 OASL CD40 JUNB SSTR2 ENST0000292729 THC2283359 CD55 IFNA4 TNFRSF21 PWWP2 CD40 KALRN TA-NFKBH ENST0000336749 THC2381535 CDC42EP2 IL18BP TNFSF10 SDC4 CD40 KCNMB1 TBX21 ENST0000344523 TNFAIP3 CLCF1 IQCD U94903 TBX21 CD40 KIAA0828 THC2266474 FAM69B TNIP1 CNR1 ISL2 UNQ5783 THC2269172 CD40 KIAA0828 THC2335499 FCER2 TRIP10 CNR1 KCNMB1 USP18 THC2314822 CD40 KIAA1549 THC2336549 FCGRT TSPAN12 CR596491 KCNMB4 ZBED3 USP18 CD40 KRT86 THC2437982 FKBP1B TTC21A CR611723 LAG3 ZBTB38 ZBTB38 CD55 LOC153346 TRIP10 FLJ10357 UPB1 CR613944 LFNG ZBTB38 CD55 LOC256021 TSPAN12 FLJ36166 USP18 CSPG2 LOC256021 ZNRF1 CDC42EP2 MAP4K4 TTC21A GADD45G ZNF358 CSRP2 LOXL3 ZNRF1 CDC42EP4 MATN4 UNC119 GAL3ST4 CYP1B1 MAP4K4 CKB MGC22014 UPB1 GPR146 CYP7B1 MLSTD1 CLIPR-59 MIA USP18 GSTO2 DGAT2 MMP15 CMKOR1 MLF1 VLDLR HSD11B1L DGAT2 MMP7 CNR1 MLF1 ZNRF1 JUNB DLL1 MOBKL2B COL1A1 MX2 ZNRF1 K03200 DUSP23 MPEG1 CR2 NEDD4L ZNRF1 KALRN CR596491 NFATC4 ZNRF1 KRT86 Features highlighted hold membership in both the training set-derived and entire dataset-derived classifiers * Certain features were represented by multiple probe-sets on the arrays and thus can be called more than once

Chapter 4: Discussion

4.1 General Discussion

The accurate identification of BRCA1/BRCA2 carriers using currently employed mutational screening technologies remains a problem in cases where the identified mutation does not result in a truncated protein. The objectives of this study were twofold; our primary objective was to determine whether BRCA1/BRCA2 mutation carriers could be distinguished from control individuals based solely on the gene expression patterns of non-cancerous (normal) cells both with and without exposure to IR, while our secondary objective was to attempt to garner some insight into the molecular mechanisms responsible for BRCA1/BRCA2 mutation carrier predisposition to cancer. The results of this study strongly suggest that both BRCA1 and BRCA2 mutation carriers are readily distinguishable from control individuals at both the baseline gene expression level, and in profiling following the induction of DNA damage. While the precise molecular mechanisms predisposing these individuals to cancer remain uncertain, the results of our study support the idea that the loss of a single copy of either of these genes results in haploinsufficiency at the cellular level resulting in deficient DNA repair and therefore a perturbed cell cycle. That the identification of mutation carriers was possible at both the baseline and IR- dependent expression profiling levels highlights the significance of the function of

BRCA1/BRCA2 in both replication-induced and exogenous insult-induced DNA damage repair.

4.2 Ionizing Radiation Induced Response in Wild Type Cells

In an effort to ensure that our selected cell type, LCLs, were undergoing an ionizing radiation response consistent with previous studies investigating the IR-response [104,105,106], 65

the expression profiles of the 38 wild type individuals were compared pre and post-treatment with

2 Gy using SAM [104]. A list of genes previously identified as top-features in these studies

[104,105,106], i.e. those that showed the greatest changes in expression, also found in our dataset are shown in Table 1. Upon comparison of the gene lists from these previous studies with that from our study we see that approximately 60-70% of previously identified genes are recapitulated in our findings. Commonly, alternate isoforms of genes previously identified were found in our study; as well, even at a stringent false discovery rate (FDR) of 1.5% a much larger number of genes (~7000, compared with 200-3000) were identified as significant in our study than in those previous. This is likely due to the use of rather limited probe sets, the largest being 10,000 genes, in these previous studies. Genes previously identified as top features and recapitulated in our study were found to occupy a wide array of functional categories including: cell cycle/cell cycle proliferation, apoptosis, DNA repair, stress response, signal transduction, RNA binding/editing, protein synthesis/degradation, energy metabolism, metabolism of macromolecular precursors and cell structure/adhesion. Also consistent with these studies was the finding that while a large percentage of the genes identified possessed functions in the cell cycle and cell cycle proliferation, apoptosis, DNA repair or the stress response—all functions previously associated with the DNA damage response—there was also a large number of genes with functions that have not been well studied in the context of the damage response. Interestingly, these studies

[104,105,106] found few changes in the expression of genes involved in the DNA DSB damage repair pathways. This finding was supported by our work, and is possibly due to the fact that many of the proteins encoded by these genes are regulated post-transcriptionally [105]. An alternative conclusion could be that the basal levels of these proteins are sufficient to deal with relatively low dose of IR used in these studies.

4.3 SVM: An efficient tool for microarray analysis

The SVM is a frequently employed tool in the analysis of microarray data as it easily extends itself to cases in which data is not linearly separable. Projection of the data into a suitable high-dimensional space is achieved through the use of a kernel function. The choice of a kernel for use in a non-linear SVM must be selected empirically by the user, and is only limited by the constraint that it must satisfy the Mercer’s condition (details are beyond the scope of our work)

[111] . The radial basis kernel works well with normally (Gaussian) distributed data [111] and hence was selected for use in this study.

4.4 Differences in gene expression observed between BRCA1/BRCA2 mutation carrier and wild type individuals at the baseline expression-profiling level are likely due to inefficiency in endogenous DNA damage repair

Studies aimed at accurately identifying BRCA1/BRCA2 mutation carriers from control individuals in normal tissues are few in number, however, the two previous studies conducted to date have concluded that no significant differences exist between BRCA1/BRCA2 mutation carriers and control individuals at the baseline gene expression level [107,108]. In contrast, the results of our study suggest that BRCA1/BRCA2 mutation carriers are readily identifiable through gene expression profiling of non-cancerous (normal) cells with a high level of accuracy; an idea supported by the findings of a study conducted by Cavalli et al. [112]. Cavalli and colleagues sought to determine whether early genomic changes associated with the BRCA1/BRCA2 heterozygous genotype were present in morphologically normal tissues acquired from prophylactic mastectomies of these mutation carriers. They found that non-cancerous tissues from

BRCA1/BRCA2mutation carriers harbor significant genetic alterations that may predispose these individuals to malignant transformation [112]. Consistent with these findings, the results of our 67

study also suggest that a global pattern of gene-expression changes not found in control individuals are present in the baseline gene expression profiles of BRCA1/BRCA2 mutation carriers.

A common explanation for BRCA1/BRCA2 mutation carrier predisposition to cancer is the idea that individuals heterozygous for either of these genes are haploinsufficient at these loci causing them to have a reduced amount of mRNA and functional protein and therefore deficient

DNA damage repair of both endogenous and exogenous origin. This deficiency in DNA damage repair leads to an increased number of chromosomal translocation and aberrations rendering the cell genetically unstable and therefore more likely to sustain further mutations and undergo neoplastic transformation [113]. Support for the idea of a haploinsufficient phenotype in

BRCA1/BRCA2 mutation carriers comes from numerous studies that observed an increased number of chromosomal rearrangements and higher rates of spontaneous sister chromatid exchange in these cells compared to wild type controls [114,115,116]. As well, a study conducted by Kim et al. [116] found that BRCA2 mutation carriers had a decreased amount of full length protein, suffered increased numbers of DSBs, and had slowed DNA repair both with and without exposure to IR. Taken together this evidence suggests the observed differences in basal gene expression between BRCA1/BRCA2 mutation carriers and wild type individuals is due to deficits in the HR repair pathway in response to stalled replication forks caused by a insufficient amount of BRCA1/BRCA2 protein in the cell.

In the absence of extrinsic DNA damaging agents chromosomal DSBs in cycling cells, and the associated chromosomal rearrangements, are thought to arise during the S-phase of the cell cycle as a result of replication across a damaged DNA template [117,118,119,120]. During

DNA replication, dsDNA become dissociated into ssDNA templates for the synthesis of two

identical sister chromatids. A lesion in one of the parental strands may cause the DNA polymerase complex to stall, potentially stalling or collapsing the replication fork [121]. The arrest forks may then be processed into DSBs which in turn are repaired through HR repair [119,122,123].

Interestingly, of the genes identified as most differentially expressed between BRCA1 or BRCA2 mutation carriers and their wild type counterparts, the only direct support for this hypothesis is the observed decrease in BRCA2 expression. The lack of identification of further genes involved in either the HR repair pathway or in DNA replication in this case may be due to the fact that the number of DSBs present in the cell in the baseline case is likely quite low compared to the levels seen following treatment with IR. Thus, with the limited sample size used in this study, any outliers present might be abrogating our ability to distinguish these subtle changes.

Upon closer inspection of the gene ontology categories of the genes identified as differentially expressed in the initial BRCA1/BRCA2 baseline classifiers commonalities can be seen. The genes identified in both classifiers largely occupy the same categories which include: cell proliferation, development and differentiation; transcription, signal transduction, cytoskeletal associated proteins, ECM remodeling and inflammation mediators. Interestingly many of these pathways are known to be altered in neoplastic transformation. The observation of a number of transcription factors (JUNB, C14orf58), as well as regulators of transcription (BTBD4,

ENST00000344523, SOX9, TBX21, ZNF358) and signal transduction (FLJ10357, KALRN,

TNFAIP3) suggests the possibility that a decrease in functional BRCA1 or BRCA2 results in the dysregulation of transcriptional control and therefore aberrant signaling. This idea is supported, at least in the case of BRCA1, by several studies. Firstly, the observation that the C-terminus of

BRCA1 modulates the phosphorylation status of the C-terminal domain of RNA pol II, negatively regulating phosphorylation by the Cdk-activating kinase (CAK), suggests that BRCA1 has the

ability to control cell cycle progression or enhance NER at sites of DNA damage [124]. As well, the observation that the BRCA1/BARD1 complex associates with, and ubiquitinates, RNA pol II resulting in its degradation and the subsequent inhibition of transcription and RNA processing lends support to this theory [125,126]. It is thought that the actions of the BRCA1/BARD1 complex in this instance work to permit access of the repair machinery to these sites of DNA damage, and therefore down-regulation of this complex would make access to, and the subsequent repair of, DNA damage difficult [125]. Finally, BRCA1 is known to act as a co-activator of transcription of NF-κB [127] and p53 [128,129], suggesting that the loss or down-regulation of

BRCA1 could result in a decrease in expression of genes under the control of these transcription factors, many of which are known to play a role in the mediation of apoptosis [128,129].

Upon closer inspection of the genes identified in the baseline classifiers (Table 4), the up- regulation JUNB in the BRCA2 baseline classifier is worth discussing as its inclusion in this list helps connect many aspects of BRCA-associated cancer. JUNB comprises one half the AP-1 transcription factor whose activation results in the up-regulation of cyclin D1 (CCND1). Recall that CCND1, once transcribed, relocates to the nucleus where it binds Cdk4/6 forming a complex that promotes the G1 to S-phase transition. As well, CCND1 has been shown to associate with the estrogen receptor alpha (ER-α) and in doing so mimics the actions of this receptor’s normal ligand, estradiol, thereby stimulating the receptor’s transcriptional targets [130]. Interestingly,

BRCA1 is known to mediate ligand-independent transcriptional repression of ER-α [130], thus taken together this evidence presents a two-pronged mechanism in which the down-regulation of

BRCA1 could promote tumorigenesis through inappropriate hormonal regulation of mammary and ovarian epithelial cell proliferation thereby providing a possible explanation for why mutations in

BRCA1 predispose mainly to breast and ovarian cancer. Finally, upon comparison of the genes

comprising the initial BRCA and BRCA2 baseline classifiers (Table 4) a large degree of overlap in features is visible. Specifically, the two classifiers share 32 genes and 4 ESTs in common suggesting that BRCA1/BRCA2 function, at least in part, in the same molecular pathways.

That the only other two studies conducted to date failed to identify significant differences in basal expression between BRCA1/BRCA2 mutation carriers and their wild type counterparts is most likely due to the sample size used in their studies. In their preliminary study, Kote-Jarai et al. [107] investigated the potential existence of a carrier expression phenotype in BRCA1 mutation carriers using a sample size of 9 mutation carriers and 8 control samples, while their second study employed a sample size of 30, one-third the size of our study, comprised of 10 BRCA1 (+/-), 10

BRCA2 (+/-) and 10 wild type samples [108]. Based on the results of our own study, as shown in the heat maps of the baseline classifiers (Fig.4/5), while overall trends in expression differences are visible, closer inspection reveals a lack of total uniformity of expression within each class.

This suggests that the detection of statistically significant differences in gene expression amongst such a limited sample size would be highly improbable, and is likely the reason for the discrepancy between the results of our study and those previously conducted.

4.5 Differences in gene expression observed between BRCA1/BRCA2 mutation carrier and wild type individuals at the radiation-dependent expression-profiling level are likely due to inefficiency in the repair of exogenous DNA Damage

The results of our study also strongly suggest that BRCA1/BRCA2 mutation carriers are readily identifiable from their wild type counterparts using gene expression profiling following the induction of DNA damage through exposure to IR. This finding is supported by the results of two previous studies [107,108] that examined the potential of gene expression profiling for

identification of BRCA1/BRCA2 mutation carriers. Interestingly, a handful of genes (7 for

BRCA2, 5 for BRCA1) (S-T4, S-T5) identified as differentially expressed at a p-value of 0.05 or less in our study were also identified as top-differentiating features in these two previous studies, suggesting that unique BRCA1/BRCA2 carrier expression-phenotypes do exist.

As in the baseline expression case, a plausible explanation for the observed differences in gene expression between mutation carriers and wild type individuals is that loss of a single copy of either BRCA1 or BRCA2 renders mutation carriers haploinsufficient at these loci, thereby leading to inefficient DNA repair resulting in a perturbed cell cycle. This idea is supported by numerous studies investigating the effects of BRCA1/BRCA2 heterozygosity on DNA damage repair following exposure to IR. Firstly, a study conducted by Buchholz and colleagues [131] found that lymphoid cells of BRCA1/BRCA2 mutation carriers showed increased numbers of chromatid breaks following exposure to IR compared with healthy control individuals. A second study conducted by Foray et al. [103] found that EBV-transformed lymphoblasts from

BRCA1/BRCA2 mutation carries exhibited increased micronucleus formation and DNA repair deficiency following treatment with IR. Further support for the idea of BRCA1/BRCA2-mediated haploinsufficieny is shown in the results of this study. Recall that in an effort to elucidate the underlying mechanisms of BRCA-mediated carcinogenesis genes significantly up or down- regulated in response to treatment with IR were identified in all three cell genotypes using SAM

[104]. Interestingly, in the case of BRCA1, a decrease in the expression of BRCA1 itself along with BARD1 and RAD51 was observed, while in the case of BRCA2, down regulation of itself and RAD51,— key players in HR repair—were observed. Compared to the observed IR-induced changes in gene expression in the wild type cell lines, the only similarities were the decreased expression of BARD1 and RAD51, however this down-regulation was significantly less than that

seen in the heterozygous cases. These findings suggest impairment in DSB DNA repair via HR is present in BRCA1/BRCA2 mutation carriers and not in wild type individuals.

Again, in order to garner insight into the molecular mechanisms underlying

BRCA1/BRCA2 mutation carrier cancer predisposition, genes differentially expressed between

BRCA1 or BRCA2 mutation carriers and their wild type counterparts at a p-value threshold of p≤0.05 were examined. As shown in S-T4, the decreased expression of a number of interesting genes, namely CDKN1B, BARD1, BRCA1, RAC1, RB1, STAT1, and CREBBP was observed. Here the observed decrease in expression of BRCA1 itself provides direct support for the idea that loss of a single copy of this gene results in a decreased amount of functional protein (i.e. haploinsufficieny). The decreased expression of RB1 and CDKN1B are also of interest as both of these genes work to maintain the G1/S-phase cell cycle checkpoint and prevent entry into S-phase

[132]. RB1 acts to sequester the E2F family of transcription factors, the products of which activate a subset of genes involved either in DNA synthesis, such as DNA polymerase alpha and proteins of the origin of replication complex, or in cell cycle control, for example cyclin E [132].

On the other hand, the CDKN1B (p27) encoded protein binds to and prevents the activation of

Cdk2/cyclin E and Cdk4/cyclin D complexes, thereby controlling progression through G1 to S- phase [57]. These cell cycle perturbing effects are likely a byproduct of a lack of functional

BRCA1, and therefore inefficient DNA damage repair. As well these changes help to explain

BRCA-mediated carcinogenesis as less stringent cell cycle checkpoints could permit cells harboring unrepaired DNA damage entry into S-phase, which if replicated could result in increased genomic instability.

The observed decreased expression of BARD1 is also of interest as BRCA1 and BARD1 are known to form a heterodimer with E3-ubiquitin ligase activity whose main target appears to

be RNA polymerase II (RNA pol II); targeting of RNA pol II by the BRCA1-BARD1 heterodimer results in its degradation and the subsequent inhibition of transcription and RNA processing (discussed above) [125]. As well, the BRCA1-BARD1 complex was recently shown to be required for ATM/ATR-mediated phosphorylation of p53 at Ser15which mediates G1/S-phase arrest following IR-induced DNA damage [57]. Again, it is possible to see how a decrease in available BRCA1 could result in less stringent cell cycle controls thereby allowing the propagation of unrepaired DNA damage into S-phase and avoidance of apoptosis. Further support for BARD1 as a potential facilitator of BRCA1-mediated cancer predisposition comes from a study conducted by Reinholz et al. [133] in which mRNA levels of BARD1 were found to be significantly reduced in invasive breast tumour tissues compared to normal tissues.

Finally, the down-regulation of CREBBP is highly interesting as it a rather promiscuous transcriptional co-activator used by a variety of transcriptional activators in a number of important signaling pathways including those of p53, Fos and Jun [134], as well as BRCA1 [135]. The cellular demand for CREBBP is so high that it is thought to be used competitively amongst these pathways [136]. If this is indeed the case, a decreased amount of CREBBP could result in the dysregulation of BRCA1-mediated transcription as well as inappropriate signaling in a number of pathways implicated in neoplastic transformation; an idea supported by the inclusion of RASSF2 and members of the MAPK-family in the final BRCA1-IR classifier.

In the case of BRCA2, inspection of genes differentially expressed between BRCA2 mutation carriers and their wild type counterparts at a threshold of p≤0.05 (S-T5), reveals the increased expression of RAD50. This increased expression of RAD50 is of particular interest as

Rad50, a member of the MRN complex, is known to tag DNA DSBs for repair via NHEJ [137].

Thus, this increased expression of RAD50 in BRCA2 mutation carriers suggests indirectly that a

decrease in functional BRCA2 results in deficient DNA repair via HR, but increased repair via

NHEJ and possibly SSA, an idea supported by several studies to date [82, 137, 83, 138].

The genes comprising the final BRCA1/BRCA2 IR-dependent classifiers (Table 4&5) are loosely categorized into the following categories: cell proliferation, development and differentiation, transcription, signal transduction, cytoskeletal associated proteins, ECM remodeling and inflammation mediators; largely the same functions seen in the baseline classifiers. Again, alterations in many of these pathways could easily contribute to neoplastic transformation. Interestingly here, comparison of the genes comprising the final IR- dependent classifier created using the training set only and those identified using the entire dataset, while showing a fair degree of overlap, show the classifier created using the training set only to lack the inclusion of two genes (BRCA1 and CREBBP) whose functions go a long way in explaining

BRCA1-mediated cancer predisposition. Perhaps not coincidentally this classifier is also far less accurate.

That membership in the final IR-dependent classifiers is more tenuous than membership in the basal classifiers makes intuitive sense, as it is possible that a number of the changes witnessed in the IR-dependent classifier are not BRCA-related, but instead are a result of exposure to IR. However, that a number of genes ranked as highly significant in the two previous BRCA carrier profiling studies conducted to date were recapitulated in our study suggests otherwise. As well, the observation that the BRCA2 IR-dependent classifiers are generally better performers than their BRCA1 IR-dependent counterparts is not surprising giving the rather specific functions of

BRCA2 compared to the pleiotrophic roles of BRCA1.

4.6 Future Directions

1. As this study is a preliminary discovery-based study, validation of the results presented here is needed. In the immediate future qPCR of genes selected in the final baseline and IR- dependent classifiers should be done for confirmation of these genes as differentially expressed.

2. From a biological standpoint, this study and others conducted to date [107,108] clearly demonstrate that the molecular mechanisms contributing to BRCA-mediated cancer predisposition are quite complex. Thus, the use of a single microarray, providing only a snap-shot of a single phase of cell cycle arrest and repair is not likely to provide much insight into these underlying pathways. If however, instead of the selection of a single time point, the assay was redone and

RNA fractions were collected sequentially at various time points following treatment, it is possible that the pathways at work would become more apparent. This idea is supported by the findings of Rieger & Chu [106] who found that changes in gene expression differed depending on the selected recovery time.

3. A certain wild type sample was consistently misclassified as cancerous by both the

BRCA1 and BRCA2 classifiers in both the basal and IR-dependent cases suggesting the possibility that this particular sample was incorrectly classified and is harboring an unidentified inactivating mutation in either BRCA1 or BRCA2 or possibly in another breast-cancer predisposition gene.

Thus, re-sequencing of this sample to ensure this is not the case is warranted.

4. The end goal of this study is the development of a novel medical diagnostic tool for the identification of hereditary breast cancer predisposing mutations, especially those in which the putative mutation does not result in a truncated protein. Thus, following confirmation of the preliminary results presented here via qPCR, the study should be transferred to a more clinically relevant system, i.e. blood. Interestingly, as the results of this study suggest that the basal changes in gene expression are sufficient for BRCA-carrier identification, treatment with IR is no longer required simplifying the study greatly.

4.7 Clinical Relevance

Breast cancer is the second leading cause of cancer-related deaths in women [1], with 5-

10% of cases being hereditary in nature [2]. Approximately 60% of all hereditary breast cancer cases are due to the presence of a mutation in either BRCA1 or BRCA2, and approximately 80% of individuals develop breast cancer by the age of 70 years [5]. Early and accurate identification of individuals with hereditary cancer risk is necessary to allow for the best possible patient outcomes. The class of mutation known as VUS remain problematic in the management of hereditary breast cancer as while current screening technologies can detect them, their designation as disease-causing or otherwise is unknown. Currently available genetic diagnostics, namely protein-truncation testing (PTT), Denaturing High Pressure Liquid Chromatography (DHPLC), and direct sequencing of coding regions, are all time and labor intensive, cost-ineffective and incapable of identifying all classes of mutations.

The preliminary results of this study suggest that all functional BRCA1 and BRCA2 mutations share a common gene expression profile indicative of their underlying molecular pathogenesis. If true, this assay possesses the possibility of development into a novel screening 77

test for hereditary breast cancer. If all classes of functional BRCA1/BRCA2 mutations do in fact posses a unique carrier expression phenotype, then any VUS that is in fact a functional mutation should display that profile, reducing the uncertainty associated with identification of a VUSs.

Also, if truly able to identify all functional mutations this assay has the possibility of identifying novel mutations, for example promoter, splicing, and intronic mutations whereby protein function is compromised by sequence changes that are not readily identifiable using currently employed screening techniques.

The molecular classifiers presented here while highly accurate did not display infallible performance, misclassifying a proportion of test samples in all cases. At this time expression- based mutation classification does not show sufficient specificity to allow it to substitute for currently in place screening tests, however, it is plausible that this technique could rapidly become an important adjunct to such techniques. If the results of this study are able to be repeated upon transference of this assay to blood, the resulting diagnostic has the potential to vastly improve the timeframe for BRCA1/BRCA2 mutation detection thereby leading to improved patient outcomes by allowing the timely initiation of patient counseling and prophylactic measures.

References

1. Canada, Canadian Cancer Society / National Cancer Institute of Canada, Canadian Cancer Statistics 2007. 2007: Toronto, Canada.

2. Deng CX. BRCA1: cell cycle checkpoint, genetic instability, DNA damage response and cancer evolution. Nucl. Acids Res. 2006; 34(5):1416-1426.

3. Stewart ZA, Pietenpol JA. Cell Cycle Checkpoints as Therapeutic Targets. J. of Mammary Gland Biology and Neoplasia. 1999;4(4):389-400.

4. Paulovich AG, Toczyski DP, Hartwell, LH. When Checkpoints Fail. Cell. 1997;88:315- 321.

5. Scully R, Puget N. BRCA1 and BRCA2 in hereditary breast cancer. Biochimie, 2002;84:95-102.

6. Boulton SJ. Cellular function of the BRCA tumour-suppressor proteins. Biochem. Soc. Trans. 2006;34:633-645.

7. Hall JM, Lee MK, Newman B, Morrow JE, Anderson LA, Huey B, King MC. Linkage of early-onset familial breast cancer to chromosome 17q21. Science. 1990;250:1684-1689.

8. Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, Liu Q, Cochran C, Bennett LM, Ding , et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science. 1994;266:66-71.

9. Stratton MR, Ford D, Neuhasen S, Seal S, Wooster R, Friedman LS, King MC, Egilsson V, Devilee P, McManus R, et al. Familial male breast cancer is not linked to the BRCA1 locus on chromosome 17q. Nat. Genet. 1994;7:103-107.

10. Wooster R, Neuhausen SL, Mangion J, Quirk Y, Ford D, Collins, N, Nguyen K, Seal S, Tran T, Averill D, et al. Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q12-13. Science. 1994;265:2088-2090.

11. Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, et al. Identification of the breast cancer susceptibility gene BRCA2 . Nature. 1995;378:789-792.

12. Breast Cancer Information Core (BIC) Database. Available at: http://research.nhgri.nih.gov/bic/. Accessed Jan, 2008.

13. Wooster R, Weber BL. Breast and ovarian cancer. N.Engl. J. Med. 2003;348:2339-2347.

14. Lorrick KL, Jensen JP, Fang S, Ong AM, Hatakeyama S, Weissman AM. RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc. Natl. Acad. Sci. USA. 1999;96:11364-11369.

15. Koonin EV, Altschul SF, Bork P. BRCA1 protein products Functional motifs. Nat. Genet. 1996;13:266-268.

16. Bork P, Hofmann K, Bucher P, Neuwald AF, Altschul SF, Koonin EV. A superfamily of conserved domains in DNA damage-responsive cell cycle checkpoint proteins. FASEB J. 1997;11:68-76.

17. Deng CX, Brodie SG. Roles of BRCA1 and its interacting proteins. Bioessays. 2000;22:728-737.

18. Venkitaraman AR. Cancer susceptibility and the functions of BRCA1 and BRCA2. Cell. 2002;108:171-182.

19. Freidman LS, Ostermeyer EA, Szabo CI, Dowd P, Lynch ED, Rowell SE, King MC. Confirmation of BRCA1 by analysis of germline mutations linked to breast and ovarian cancer in ten families. Nat. Genet. 1994;8:399-404.

20. Kinzler KW, Vogelstein B. Cancer-susceptibility genes. Gatekeepers and caretakers. Nature, 1997;386:761-763.

21. Bork P, Blomberg N, Nilges M. Internal repeats in the BRCA2 protein sequence. Nat. Genet. 1996;13:22-23.

22. Wong AK, Pero R, Ormonde PA, Tavtigian SV, Bartel PL. Rad51 interacts with the evolutionarily conserved BRC motifs in the human breast cancer susceptibility gene BRCA2. J. Biol. Chem. 1997;272:31941-31944.

23. Davies AA, Masson JY, Mellwraith MJ, Stasiak AZ, Venkitaraman AR, West SC. Role of BRCA2 in control of the RAD51 recombination and DNA repair protein. Mol. Cell. 2001;7:273-282.

24. Gundmundsdottir K, Ashworth A. The roles of BRCA1 and BRCA2 and associated proteins in the maintenance of genomic stability. Oncogene, 2006;25:5864-5874.

25. Smith SA, Easton DF, Evans DG, Ponder BA. Allele losses in the region 17q12–21 in familial breast and ovarian cancer involve the wild-type chromosome. Nat. Genet. 1992;2:128–131.

26. Collins N, McManus R, Wooster R, Mangion J, Seal S, Lakhani SR et al. Consistent loss of the wild type allele in breast cancers from a family linked to the BRCA2 gene on chromosome 13q12–13. Oncogene. 1995;10:1673–1675.

27. Hartwell LH, Weinert TA. Checkpoints: controls that ensure the order of cell cycle events. Science. 1989;246:629-634.

28. Sherr CJ. G1 Phase progression: Cycling on cue. Cell. 1994;79:551-555.

29. Ekholm SV, Reed SI. Regulation of the G1cyclin-dependent kinases in the mammalian cell cycle. Curr Opin Cell Biol. 2000;12(6):676-684.

30. Grana X, Reddy EP. Cell cycle control in mammalian cells: role of cyclins, cyclin dependent kinases (CDKs), growth suppressor genes and cyclin-dependent kinase inhibitors (CKIs). Oncogene. 1995;11(2):211-219.

31. Loeb LA. 1991. Mutator phenotype may be required for multistage carcinogenesis. Cancer Res. 1991;51:3075– 3079.

32. Nyberg KA, Michelson RJ, Putnam CW, Weinert TA. Toward Maintaining The Genome: DNA damage and replication checkpoints. Annu. Rev. Genet. 2002;36:617-656.

33. Hartwell L, Mortimer RK, Culotti J, Culotti M. Genetic control of the cell division cycle in yeast: genetic analysis of cdc mutants. Genetics. 1973;74:267-286.

34. Abraham RT. Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes Dev. 2001;15:2177-2196.

35. Durocher D, Jackson SP. DNA-PK, ATM and ATR, variations on a theme? Curr. Opin. Cell Biol. 2001;2:225–231.

36. Melo J, Toczyski D. A unified view of the DNA-damage checkpoint. Curr. Opin. Cell Biol. 2002;14:237–245.

37. Banin S, Moyal L, Shieh S, Taya Y, Anderson SW, Chessa L, et al. Enhanced phosphorylation of p53 by ATM in response to DNA damage. Science. 1998;281:1674- 1677.

38. Ünsal-Kacmaz K, Makhov AM, Griffith JD, Sancar A. Preferential binding of ATR protein to UV Damaged DNA. Proc. Natl. Acad. Sci. 2002;99:6673-6678.

39. Schultz LB, Chehab NH, Malikzay A, Halazonetis T. p53 binding protein 1 (53BP1) is an early participant in the cellula response to DNA double-strand breaks. J. Biol. Chem. 2000;151:1381-1390.

40. Yamane K, Wu XL, Chen J. A DNA damage-regulated BRCT-containing protein, TopBP1, is required for cell survival. Mol. Cell. Biol. 2002;22:555-566.

41. Goldberg M, Stucki M, Falck J, D’Amours D, Rahman D, Pappin D, et al. MDC1 is required for the intra S-phase DNA damage checkpoint. Nature. 2003;421:952-956.

42. Wang Y, Cortez D, Yazdi P, Neff N, Elledge S, Qin J. BASC, A super complex of BRCA1-associated proteins involved in the recognition and repair of aberrant DNA structures. Genes Dev. 2000;14:927-939.

43. Rhind N, Russell P. Chk1 and Cds1: linchpins of the DNA damage and replication checkpoint pathways. J. Cell. Science. 2000;113:3889-3896.

44. Melo JA, Cohen J, Toczyski DP. Two checkpoint complexes are independently recruited to sites of DNA damage in vivo. Genes Dev. 2001;15:2809-2821.

45. Matsuoka S, Rotman G, Ogawa A, Shiloh Y, Tamai K, Elledge SJ. Ataxia telangiectasia- mutated phosphorylates Chk2 in vivo and in vitro. Proc. Natl. Acad. Sci. 2000;97:10389- 10394.

46. Zhoa H, Piwnica-Worms H. ATR-mediated checkpoint pathways regulate phosphorylation an activation of human Chk1. Mol. Cell. Biol., 2001;21:4129-4139.

47. Bartek J, Lukas J. Mammalian G1- and S-phase checkpoints in response to DNA damage. Curr. Opin. Cell Biol. 2001;13:738–747.

48. Mailand N, Falck J, Lukas C, Syljuasen RG, Welcker M, et al. Rapid destruction of human Cdc25A in response to DNA damage. Science. 2000; 288:1425– 1429.

49. Canman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, et al. Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science. 1998;281: 1677– 79.

50. Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev. 1999;13:1501– 1512.

51. Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ. The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell. 1993;75: 805– 816.

52. Paulovich AG, Hartwell LH. A checkpoint regulates the rate of progression through S phase in S. cerevisiae in response to DNA damage. Cell. 1995;82: 841– 847.

53. Gould K, Nurse P. Tyrosine phosphorylation of the fission yeast cdc2+ protein kinase regulates entry into mitosis. Nature. 1989;342:39-45.

54. Yarden RI, Pardo-Reoyo S, Sgagias M, Cowan KH, Brody LC. BRCA1 regulates the G2/M checkpoint by activating Chk1 kinase upon DNA damage. Nat. Genet. 2002;30:285– 289.

55. Sanchez Y, Wong C, Thoma RS, Richman R, Wu Z, Piwnica-Worms H, Elledge S. Conservation of the Chk1 checkpoint pathway in mammals: linkage of DNA damage to Cdk regulation through Cdc25. Science. 1997;277:1497-1501.

56. Somasundaram K, Zhang H, Zeng YX, Houvras Y, Peng Y, Wu GS, et al. Arrest of the cell cycle by the tumour-suppressor BRCA1 requires the CDK-inhibitor p21WAF1/CiP1. Nature. 1997;389:187–190.

57. Fabbro M, Savage K, Hobson K, Deans AJ, Powell SN, McArthur GA, Khanna KK. BRCA1-BARD1 complexes are required for p53Ser-15 phosphorylation and a G1/S arrest following ionizing radiation-induced DNA damage. J. Biol. Chem. 2004;279:31251–31258.

58. Aprelikova ON, Fang BS, Meissner EG, Cotter S, Campbell M, Kuthiala A, Bessho M, Jensen RA, Liu ET. BRCA1-associated growth arrest is RB-dependent. Proc. Natl Acad. Sci. USA. 1999;96:11866–11871.

59. Xu B, Kim S, Kastan MB. Involvement of Brca1 in S-phase and G(2)-phase checkpoints after ionizing irradiation. Mol. Cell Biol. 2001;21:3445–3450.

60. Hutchins JR, Clarke PR. Many fingers on the mitotic trigger: post-translational regulation of the Cdc25C phosphatase. Cell Cycle, 2004;3:41–45.

61. Mullan PB, Quinn JE, Gilmore PM, McWilliams S, Andrews H, Gervin C, et al. BRCA1 and GADD45 mediated G2/M cell cycle arrest in response to antimicrotubule agents. Oncogene. 2001;20:6123-6131.

62. Friedberg EC, Walker GC, Siede W. DNA repair and mutagenesis. ASM Press, Washington, D.C.; 1995.

63. Ward JF. DNA damage produced by ionizing radiation in mammalian cells: identities, mechanisms of formation, and reparability. PNAS. 1988;35:95-125.

64. Kanaar R, Hoeijmakers HJ, van Gent DC. Molecular mechanisms of DNA double-strand break repair. Trends Cell Biol. 199;8(12):483-489.

65. Ivanov EL, Haber JE. DNA repair: RAD alert. Curr. Biol. 1997;7:492-495.

66. Jeggo PA. DNA breakage and repair. Adv Genet. 1998;38:185-218.

67. West SC, Chappell C, Hanakahi LA, Masson JY, McIlwraith MJ, Van Dyck E. Double- strand break repair in human cells. Cold Spring Harb Symp Quant Biol. 2000;65:315-21.

68. Lees-Miller SP, Meek K. Repair of DNA double strand breaks by non-homologous end joining. Biochimie, 2003;85(11):1161-1173.

69. Zhang J, Willers H, Feng Z, Ghosh JC, Kim S, Weaver DT, et al. Chk2 phosphorylation of BRCA1 regulates DNA double-strand break repair. Mol Cell Biol. 2004;24(2):708-18.

70. Mirzoeva OK, JH Petrini. DNA damage-dependent nuclear dynamics of the Mre11 complex. Mol Cell Biol. 2001; 21(1):281-288.

71. van den Bosch M, Bree RT, Lowndes NF. The MRN complex: coordinating and mediating the response to broken chromosomes. EMBO Rep. 2003;4(9):844-849.

72. Paull TT, Gellert M. Nbs1 potentiates ATP-driven DNA unwinding and endonuclease cleavage by the Mre11/Rad50 complex. Genes Dev. 1999;13:1276–1288.

73. Lieber MR, Ma Y, Pannicke U, Schwarz K. Mechanism and regulation of human non- homologous DNA end-joining. Nat Rev Mol Cell Biol. 2003;4(9):712-720.

74. Stark JM, Pierce AJ, Oh J, Patink A, Jasin M. Genetic steps of mammalian homologous repair with distinct mutagenic consequences. Mol Cell Biol. 2004;24(21):9305-9316.

75. Smith GC, Jackson SP. The DNA-dependent protein kinase. Genes Dev. 1999;13:916- 934.

76. Budman J, Kim SA Chu G. Processing of DNA for nonhomologous end-joining is controlled by kinase activity and XRCC4/ligase IV. J Biol Chem. 2007;282(16):11950- 11959.

77. Rothkamm K, Kruger I, Thompson LH, Lobrich M. Pathways of DNA double-strand break repair during the mammalian cell cycle. Mol Cell Biol. 2003;23(16):5706-5715.

78. West SC. Molecular views of recombination proteins and their control. Nat Rev Mol Cell Biol. 2003;4(6):435-445.

79. Liu Y, et al. Rad51 is required for Holliday junction processing in mammalian cells. Science. 2004;303(5655):243-246.

80. Elledge SJ. Cell cycle checkpoints: preventing an identity crisis. Science. 1996;274:1664- 1672.

81. Kastan MB, Bartek J. Cell cycle checkpoints and cancer. Nature. 2004;432:316-323.

82. Moynahan ME, Pierce AJ, Jasin M. BRCA2 is required for homology-directed repair of chromosomal breaks. Mol Cell 2001;7:263–272.

83. Patel KJ, Vu VP, Lee H, Corcoran A, Thistlethwaite FC, Evans MJ, et al. Involvement of Brca2 in DNA repair. Mol Cell. 1998;1: 347–357.

84. Deng CX, Wang RH. Roles of BRCA1 in DNA damage repair: a link between development and cancer. Hum Mol Genet. 2003;Spec No 1:R113-23.

85. Moynahan ME, Chiu JW, Koller BH, Jasin M. Brca1 controls homology-directed DNA repair. Mol Cell. 1999;4(4):511-8.

86. Zhang J, Powell SN. The role of the BRCA1 tumor suppressor in DNA double-strand break repair. Mol Cancer Res. 2005;3(10):531-539.

87. Powell SN, Willer H, Xia F. BRCA2 keeps Rad51 in line. High-fidelity homologous recombination prevents breast and ovarian cancer? Mol Cell. 2002;10:1262-3.

88. Valerie K, Povirk LF. Regulation and mechanisms of mammalian double-strand break repair. Oncogene 2003;22:5792-812.

89. Haber, JE. The many interfaces of Mre11. Cell. 1998;95:583-586.

90. Paull TT, Rogakou RP, Yamazaki V, Kirchgessner CU, Gellert M, Bonner WM. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Bio.l 2000;10:886-895.

91. Scully R, Chen J, Plug A, Xiao Y, Weaver D, Feunteun J, Ashley T, Livingston DM. Association of BRCA1 with Rad51 in mitotic and meiotic cells. Cell. 1997;88:265-75.

92. Yu VP, Koehler M, Steinlein C, Schmid M, Hanakahi LA, van Gool AJ, et al. Gross chromosomal rearrangements and genetic exchange between nonhomologous chromosomes following BRCA2 inactivation. Genes Dev. 2000;14:1400-1406.

93. Yoshida K, Miki Y. Role of BRCA1 and BRCA2 as regulators of DNA repair, transcription, and cell cycle in response to DNA damage. Cancer Sci. 2004;95:866-871.

94. Stekel D. Microarrray Bioinformatics. New York, NY: Cambridge University Press; 2003.

95. Lee MT. Analysis of microarray gene expression data. Boston, MA: Kluwer Academic; 2004.

96. Quackenbush J. Microarray data normalization and transformation. Nat. Genet. 2002;32(S2):496-501.

97. Yeung KY, Bumgarner RE. Multiclass classification of microarray data with repeated measurements: application to cancer. Genome Biol. 2003;4:R83.

98. Boser BE, Guyon IM, Vapnik VN. A training algorithm for optimal margin classifiers. In: D. Haussler, editor, 5th Annual ACM Workshop on COLT, Pittsburgh, PA, ACM Press; 1992:144-152.

99. Vapnik VN. Statistical learning theory. New York, NY: Wiley-Interscience; 1998.

100. Tang Y, Zhang YQ, Huang Z. Development of two-stage SVM-RFE gene selection strategy for microarray expression data analysis. IEEE/ACM Trans Comput Biol Bioinform. 2007;4(3):365-81.

101. Rbf_SVC algorithm. Dr. Saeed Hashimi Homepage. Available at: http://torch.cs.dal.ca/~saeed/index.php?option=com_content&task=blogcategory&id=34 &Itemid=64. Accessed Oct, 2007.

102. Sturn A, Quackenbush J, Trajanoski Z. Genesis: cluster analysis of microarray data. Bioinformatics. 2002;18:207–8.

103. Foray N, Randrianarison D, Marot D, Perricaudet M, Lenoir G, Feunteun J. Gamma-rays- induced death of human cells carrying mutations of BRCA1 or BRCA2. Oncogene. 1999; 18 (51):7334–7342.

104. Tusher V, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. 2001;98:5116-5121.

105. Akerman GS, Rosenzweig BA, Domon OE, Tsai CA, Bishop ME, McGarrity LJ, Macgregor JT, Sistare FD, Chen JJ, Morris SM. Alterations in gene expression profiles and the DNA-damage response in ionizing radiation-exposed TK6 cells. Environ Mol Mutagen. 2005;45(2-3):188-205.

106. Rieger KE, Chu G. Portrait of transcriptional responses to ultraviolet and ionizing radiation in human cells. Nucleic Acids Res. 2004;32(16):4786-803.

107. Kote-Jarai Z, Williams RD, Cattini N, Copeland M, Giddings I, Wooster R, et al. Gene expression profiling after radiation-induced DNA damage is strongly predictive of BRCA1 mutation carrier status. Clin Cancer Res. 2004;10(3):958-63.

108. Kote-Jarai Z, Matthews L, Osorio A, Shanley S, Giddings I, Moreews F, et al. Accurate prediction of BRCA1 and BRCA2 heterozygous genotype using expression profiling after induced DNA damage. Clin Cancer Res. 2006;12(13):3896-901.

109. Ramaswamy S, Tamayo P, Rifkin R, Mukherjee S, Yeang CH, Angelo M, et al. Multiclass cancer diagnosis using tumor gene expression signatures. Proc Natl Acad Sci U S A. 2001;98(26):15149-54

110. Stevens R, Goble CA, Bechhofer S. Ontology-based knowledge representation for bioinformatics. Brief Bioinform. 2000;1:398-414.

111. Vapnik, V. The Nature of Statistical Learning Theory. New York, NY: Springer-Verlag; 1995.

112. Cavalli LR, Singh B, Isaacs C, Dickson RB, Haddad BR. Loss of heterozygosity in normal breast epithelial tissue and benign breast lesions in BRCA1/2 carriers with breast cancer. Cancer Genet Cytogenet. 2004;149(1):38-43.

113. Santarosa M, Ashworth A. Haploinsufficiency for tumour suppressor genes: when you don't need to go all the way. BBA. 2004;2:105-122.

114. Savelyeva L, Claas A, Matzner I, Schlag P, Hofmann W, Scherneck S, et al. Constitutional genomic instability with inversions, duplications, and amplifications in 9p23–24 in BRCA2 mutation carriers. Cancer Res. 2001;61(13): 5179–5185.

115. Schwab M, Claas A, Savelyeva L. BRCA2: a genetic risk factor for breast cancer. Cancer Lett. 2002;175(1):1–8.

116. Kim MK, Zitzmann S, Westermann F, Arnold K, Brouwers S, Schwab M, et al. Increased rates of spontaneous sister chromatid exchange in lymphocytes of BRCA2+/- carriers of familial breast cancer clusters. Cancer Lett. 2004;210(1): 85–94.

117. Kowalczykowski SC. Initiation of genetic recombination and recombination-dependent replication. Trends Biochem. Sci. 2000;25: 156–165.

118. Cox MM, Goodman MF, Kreuzer KN, Sherratt DJ, Sandler SJ, Marians KJ. The importance of repairing stalled replication forks. Nature2000;404:37–41.

119. Michel B, Grompone G, Flores MJ, Bidnenko V. Multiple pathways process stalled replication forks. Proc. Natl. Acad. Sci. USA 2004;101:12783–12788.

120. Chen C, Kolodner RD. Gross chromosomal rearrangements in Saccharomyces cerevisiae replication and recombination defective mutants. Nat. Genet. 1999;23:81–85.

121. Nagaraju G, Scully R. Minding the gap: the underground functions of BRCA1 and BRCA2 at stalled replication forks. DNA Repair (Amst). 2007;6(7):1018-31.

122. Takata M, Sasaki MS, Sonoda E, Morrison C, Hashimoto M, Utsumi H, et al. Homologous recombination and non-homologous end-joining pathways of DNA double- strand break repair have overlapping roles in the maintenance of chromosomal integrity in vertebrate cells. EMBO J. 1998;17:5497–5508.

123. Helleday T. Pathways for mitotic homologous recombination in mammalian cells, Mutat. Res. 2003;532:103–115.

124. Moisan A, Larochelle C, Guillemette B, Gaudreau L. BRCA1 can modulate RNA polymerase II carboxy-terminal domain phosphorylation levels. Mol Cell Biol. 2004 ;24: 6947–6956.

125. Kleiman FE, Wu-Baer F, Fonseca D, Kaneko S, Baer R, Manley JL. BRCA1/BARD1 inhibition of mRNA 3' processing involves targeted degradation of RNA polymerase II. Genes Dev. 2005;19:1227–1237.

126. Starita LM, Horwitz AA, Keogh MC, Ishioka C, Parvin JD, Chiba N. BRCA1/BARD1 ubiquitinate phosphorylated RNA polymerase II. J Biol Chem. 2005;280:24498–24505.

127. Houvras Y, Benezra M, Zhang H, Manfredi J J, Weber BL, Licht JD. BRCA1 Augments Transcription by the NF-κB Transcription Factor by Binding to the Rel Domain of the p65/RelA Subunit. J. Biol. Chem. 2000;275:36230–36237.

128. Ouchi T, Monteiro AN, August A, Aaronson SA, and Hanafusa H. BRCA1 regulates p53- dependent gene expression. Proc. Natl. Acad. Sci. U. S. A. 1998;95:2302–2306.

129. Zhang, H., Somasundaram, K., Peng, Y., Tian, H., Bi, D., Weber, B. L., and El-Deiry, W. S. BRCA1 physically associates with p53 and stimulates its transcriptional activity. (1998) Oncogene 16, 1713–1721.

130. Zheng L, Annab LA, Afshari CA, Lee WH, Boyer TG. BRCA1 mediates ligand- independent transcriptional repression of the estrogen receptor. Proc Natl Acad Sci USA 2001;98: 9587–9592.

131. Buchholz TA, Wu X, Hussain A, Tucker AL, Mills GB, Haffty B, et al. Evidence of haplotype insufficiency in human cells containing a germline mutation in BRCA1 or BRCA2. Int. J. Cancer. 2002; 97 (5): 557–561.

132. DeGregori J, Kowalik T, Nevins JR. Cellular targets for activation by the E2F1 transcription factor include DNA synthesis- and G1/S-regulatory genes. Mol Cell Biol. 2001;15:4215-4224.

133. Reinholz MM, An MW, Johnsen SA, Subramaniam M, Suman VJ, Ingle JN, Roche PC, Spelsberg TC. Differential gene expression of TGF beta inducible early gene (TIEG), Smad7, Smad2 and Bard1 in normal and malignant breast tissue. Breast Cancer Res Treat. 2004; 86(1):75-88.

134. Janknecht R, Hunter T. Transcription. A growing coactivator network. Nature.1996;383:22-23.

135. Pao GM, Janknecht R, Ruffner H, Hunter T, Verma IM. CBP/p300 interact with and function as transcriptional coactivators of BRCA1. Proc Natl Acad Sci U S A. 2000;97(3):1020-5.

136. Yao TP, Oh SP, Fuchs M, Zhou ND, Ch'ng LE, Newsome D, Bronson RT, Li E, Livingston DM, Eckner R. Gene dosage-dependent embryonic development and proliferation defects in mice lacking the transcriptional integrator p300. Cell.1998;93: 361-372.

137. Paull TT, Cortez D, Bowers B, Elledge SJ, Gellert M. Direct DNA binding by Brca1. Proc Natl Acad Sci USA. 2001;98: 6086–6091.

138. Larminat F, Germanier M, Papouli E, Defais M. Deficiency in BRCA2 leads to increase in non-conservative homologous recombination. Oncogene. 2002;21:5188–5192.

Appendix A: Supplementary Tables

S-T1: Changes in Gene Expression of Wild Type Cells following Exposure to IR….92 S-T2: BRCA1-IR Classifier (training set only) Features at p≤0.05: 1110 Features….105 S-T3: BRCA2-IR Classifier (training set only) Features at p≤0.05: 476 Features…...110 S-T4: BRCA1-IR-AD Classifier Features at p≤0.05: 907 Features………………….112 S-T5: BRCA2-IR-AD Classifier Features at p≤0.05: 503 Features…………………..116 S-T6: BRCA1-IR Response-SAM-6158 Features……………………………………118 S-T7: BRCA2-IR Response- SAM-4326 Features…………………………………...131

TABLE ST-1: WILD TYPE IR-RESPONSE GENES: UP-REGULATED GENES BY SAM SCORE (2643)* Gene ID Score Gene ID Score Gene ID Score Gene ID Score SULF2 8.94 MAMDC4 4.76 HGS 3.98 PPM1B 3.45 THC2429167 8.49 THC2316748 4.74 CP110 3.97 CREB3L1 3.44 ASTN2 8.22 SPATA18 4.74 THC2376737 3.94 BTBD14A 3.43 A_32_P52153 8.14 THC2340838 4.73 PODXL 3.92 CCDC92 3.43 HES2 7.69 PGAP1 4.70 JUND 3.91 SLC26A11 3.42 GIPR 7.58 CNGB1 4.66 BLOC1S2 3.89 FAM98C 3.42 THC2437069 7.57 GDF15 4.63 ENST00000235345 3.89 AHRR 3.41 SESN1 7.49 CDKN1A 4.59 OR11A1 3.87 LOC257396 3.41 VWCE 7.47 BTBD14B 4.59 PDGFRA 3.87 ZNFN1A4 3.40 AK024898 7.29 TM7SF3 4.53 SLC7A6 3.87 ZMYM1 3.40 MGC5370 7.11 A_32_P12282 4.52 DLGAP4 3.86 THC2274334 3.40 GLS2 7.06 SESN2 4.51 FLJ11259 3.86 ENST00000294485 3.40 ANKRD47 7.03 THC2439499 4.50 FAM41C 3.84 SLC35D1 3.39 C1orf183 7.02 DIRC1 4.49 CAND1 3.84 GRIN2C 3.39 ENST00000377836 6.99 PLXNB2 4.45 DCP1B 3.83 SCGB1A1 3.38 PLK2 6.95 ARHGEF3 4.44 THC2266610 3.82 DHTKD1 3.38 TP53I3 6.81 AK056245 4.42 THC2280109 3.82 LOC286254 3.38 TNFRSF10C 6.77 A_23_P64962 4.40 ITPR2 3.81 LOC57149 3.38 BC040303 6.67 BI029121 4.39 THC2238625 3.81 ADRB2 3.38 LIF 6.52 LOC653374 4.38 GLT8D3 3.81 PSTPIP2 3.37 SMPD3 6.42 A_23_P206568 4.37 TNFRSF10B 3.80 KIAA1219 3.37 ZNF385 6.25 LOC653483 4.36 FAS 3.80 MAGEL2 3.36 THC2343350 6.17 RRM2B 4.36 AA887631 3.78 MDS025 3.36 AI500335 6.17 GLTSCR1 4.36 CR627133 3.77 C14orf28 3.35 BBC3 6.12 DB518505 4.33 DUSP18 3.77 CB250445 3.35 AK092083 6.06 A_24_P144487 4.30 ENST00000360523 3.74 NSF 3.34 TRIM22 6.01 BF960555 4.29 RRAD 3.74 THC2450504 3.34 TRIAP1 5.92 FKSG2 4.26 PROCR 3.73 OR4D2 3.33 RHO 5.92 BC053363 4.24 LOC92017 3.71 A_32_P108748 3.33 C12orf5 5.84 CEACAM1 4.23 PRKAB1 3.70 FAM84A 3.32 THC2315966 5.74 STARD4 4.23 TAAR5 3.70 MAP3K7IP3 3.31 SLC6A19 5.70 FLJ13576 4.23 ZNF79 3.68 KRTAP5-8 3.31 GRHL3 5.70 C3orf23 4.22 KIAA0284 3.66 FAM46A 3.31 CN430223 5.60 THC2289056 4.21 DCUN1D3 3.66 OR10H2 3.31 FHL2 5.58 FUCA1 4.20 PSKH1 3.65 A_24_P229728 3.31 AK026338 5.58 A_24_P531074 4.20 ARHGEF10L 3.65 BC035180 3.30 NTN1 5.47 AK026194 4.17 NDFIP2 3.64 THC2408506 3.29 CABYR 5.46 XPC 4.14 IGSF9 3.63 A_32_P139021 3.29 FOSL1 5.45 FEZ1 4.14 THC2322041 3.62 TRIP6 3.28 LOC134147 5.45 SPARC 4.14 ZNF167 3.62 SGPL1 3.28 PPM1D 5.33 TNFSF4 4.13 LOC441245 3.61 PHACTR4 3.28 DOCK4 5.32 MOSPD1 4.12 CD274 3.59 KCTD1 3.27 A_23_P6514 5.30 SLC30A3 4.11 SFXN5 3.59 DQX1 3.27 SARDH 5.24 C2orf13 4.11 LOC339768 3.59 USP33 3.27 IL10RB 5.23 THC2346713 4.11 CHST6 3.58 ADAM10 3.27 ACTA2 5.18 FOLR2 4.10 TRIM35 3.57 ADH5 3.27 CFD 5.14 C1orf42 4.10 BAIAP2L1 3.57 CR603195 3.26 AF144054 5.13 ITGAM 4.09 OR7E13P 3.56 BC030100 3.26 ITPKC 5.13 TMEM35 4.08 ATF3 3.56 PHF20L1 3.25 SMTN 5.11 C1orf57 4.08 THC2406514 3.56 SUCLA2 3.25 PHLDA3 5.08 C20orf161 4.08 SLC13A2 3.55 BCAS2 3.24 ENST00000340381 5.04 THC2314369 4.07 KRT17 3.52 RAB8B 3.24 C8orf38 5.01 ISG20L1 4.05 ANKRA2 3.50 MDM2 3.24 THC2273762 4.99 SDC1 4.04 LENG8 3.50 IL27 3.24 FAM83H 4.96 LRDD 4.04 C16orf5 3.48 ALOX5 3.23 PRSS36 4.94 MLLT1 4.03 SLC4A11 3.47 EFNB1 3.23 GADD45A 4.93 A_32_P233569 4.03 THC2277837 3.46 MGC70863 3.23 ARHGEF15 4.87 BRMS1L 4.00 FBXO22 3.45 MARVELD3 3.22 WDR63 4.79 OAT 3.99 SLFN5 3.45 EI24 3.22 *ONLY GENES WITH SAM SCORE OF 2 OR GREATER ARE SHOWN

TABLE ST-1 CONT’D: UP-REGULATED Gene ID Score Gene ID Score Gene ID Score Gene ID Score SYTL1 3.21 TSPYL3 3.05 DR1 2.87 3.8-1 2.72 GFAP 3.21 SLCO2B1 3.04 NIPSNAP3A 2.86 AK022339 2.72 LOC253981 3.21 SGOL1 3.04 C19orf46 2.86 THC2257370 2.71 OSRF 3.21 CR590071 3.02 BC066984 2.85 LOC158863 2.71 HNRPH2 3.21 UBQLN1 3.01 BX101252 2.85 MAB21L2 2.70 Ells1 3.21 AK098185 3.01 GFI1B 2.85 SF3A1 2.70 AKAP7 3.20 AK123704 3.01 CD242823 2.85 A_24_P916853 2.70 ZNF425 3.19 FKSG44 3.00 ARL2BP 2.85 NDRG4 2.70 THC2436415 3.19 EEA1 3.00 ENST00000368491 2.84 LOC51136 2.70 TCP11L1 3.19 PLK3 3.00 AF086139 2.84 BX093417 2.70 CCNG1 3.19 METTL7A 3.00 FAM49B 2.84 PIGR 2.69 THC2243803 3.18 C18orf56 3.00 ULK1 2.84 PHF15 2.69 ZBTB20 3.18 ARHGAP6 3.00 PSMD10 2.83 KIAA0999 2.68 PTMS 3.18 TEAD3 3.00 ZNF469 2.83 ZNF35 2.68 MICB 3.17 MFSD4 2.99 C1orf26 2.82 PRKX 2.68 NIN 3.16 LOC647090 2.99 THC2345956 2.82 HTR7P 2.68 IBRDC3 3.15 LPHN1 2.99 HSDL2 2.82 PTP4A1 2.68 STK4 3.15 HIF1A 2.99 ENST00000265450 2.82 CR603184 2.68 TMPRSS2 3.15 GRAMD3 2.99 ZNF746 2.82 ARL6IP 2.68 CES2 3.15 MAD1L1 2.98 C20orf11 2.81 WFS1 2.68 EMX1 3.15 CR609588 2.98 LMBR1 2.81 THC2279548 2.68 COL4A1 3.14 KIAA0323 2.98 MYH10 2.80 THC2440818 2.67 PTPN1 3.14 PIGH 2.97 ENST00000369739 2.80 FPGT 2.67 RRAGA 3.14 LCE1F 2.97 A_23_P207049 2.80 B3GNT8 2.67 PHF6 3.14 GPR172B 2.96 CRHR1 2.80 CRYZL1 2.67 CDC42EP4 3.13 ZNF650 2.96 RAB5A 2.80 C17orf85 2.67 SNX22 3.13 ANKRD46 2.96 RCCD1 2.79 FLJ39370 2.67 PLA2G4D 3.13 SPTY2D1 2.96 LRRIQ2 2.79 C3orf50 2.66 ENST00000373335 3.12 RAB43 2.96 CR606637 2.79 KRTAP10-10 2.66 RBL2 3.12 SIDT2 2.95 SMAD5 2.79 APBA3 2.66 C1orf102 3.12 ITM2A 2.95 MYO5A 2.78 LIPA 2.66 NCSTN 3.11 A_24_P127042 2.94 TP53INP1 2.78 ABHD9 2.66 ENST00000301807 3.11 MAP4K4 2.94 NDUFV3 2.78 FBXO28 2.66 LONPL 3.11 LOC645431 2.94 PNMA3 2.78 A_24_P281853 2.65 NAP1L5 3.11 CASC4 2.94 SRA1 2.77 RABGGTA 2.65 RASD1 3.11 TNFRSF1A 2.94 OR8B8 2.77 DEPDC6 2.64 ENST00000290607 3.10 ABCA12 2.94 OASL 2.77 BM989484 2.64 A_23_P17152 3.10 PGPEP1 2.93 AF336795 2.77 FLJ32255 2.63 A_32_P30187 3.10 STOX2 2.93 DIRC2 2.77 APBB3 2.63 FCER1G 3.09 SPAST 2.93 MMP28 2.77 APOL2 2.63 S75896 3.09 AADACL1 2.92 CCDC55 2.77 CYP3A5 2.63 PCGF5 3.09 TANC1 2.92 ZNF337 2.77 THC2443137 2.62 LOC133874 3.09 CYB5R1 2.92 ARMCX6 2.77 CKAP2 2.62 ARHGAP23 3.08 CCDC113 2.92 A_24_P144625 2.76 THC2374442 2.62 CFL2 3.07 BNIP2 2.91 MRRF 2.76 SLC43A2 2.62 AY007156 3.07 NPAL3 2.91 FGF3 2.76 PRIMA1 2.62 LOC643513 3.07 MRPL49 2.91 C20orf12 2.76 THC2366161 2.61 A_32_P166152 3.07 C19orf6 2.91 ENST00000277575 2.76 HERC5 2.61 C12orf49 3.06 GYG1 2.91 C1orf88 2.76 OR7E47P 2.61 ENST00000378179 3.06 PPFIBP2 2.90 CNR1 2.76 MT 2.61 AF086329 3.06 CABC1 2.90 HHAT 2.76 RAB40B 2.61 MT3 3.06 PDK2 2.89 LCE2A 2.75 PCDHGA8 2.61 ELF5 3.06 CCR3 2.89 P2RY10 2.74 VHL 2.61 GPR109B 3.05 LYST 2.89 TNFRSF10A 2.74 C1orf144 2.60 MECP2 3.05 CXorf43 2.88 ZFP41 2.74 HOXB6 2.60 OR2H1 3.05 CASK 2.88 RBM41 2.74 FLJ12688 2.60 CSNK1D 3.05 THC2310998 2.88 SCRN3 2.73 ENST00000367233 2.60 AW977527 3.05 A_32_P172002 2.87 CASC5 2.72 A_32_P40375 2.60 SIDT1 3.05 RETSAT 2.87 SART2 2.72 PCNP 2.60

TABLE ST -1 CONT’D: UPREGULATED GENES BY SAM SCORE Gene ID Score Gene ID Score Gene ID Score Gene ID Score C6orf27 2.60 AK093416 2.50 A_24_P127462 2.41 TUBB4 2.34 TSG101 2.60 BF718543 2.50 SNAPC5 2.41 HEAB 2.34 RNF141 2.59 MAWBP 2.50 SLA/LP 2.41 USP6NL 2.34 NDUFA5 2.59 FLJ21657 2.50 FRK 2.41 POLR2K 2.33 ABCD3 2.59 THC2439430 2.50 C19orf12 2.41 SLC25A20 2.33 HCFC2 2.59 TMEM129 2.50 YES1 2.41 RP11-78J21.1 2.33 DMD 2.59 CCAR1 2.49 SLC39A13 2.40 CCNL2 2.33 D4ST1 2.59 ENST00000317868 2.49 KIAA1005 2.40 MIB2 2.33 GHSR 2.58 DEPDC5 2.49 SLTM 2.40 ENST00000361453 2.33 AL390181 2.58 BX111592 2.49 ENST00000367590 2.40 HMGB3 2.33 LOC158960 2.58 A_23_P46070 2.49 SASH1 2.40 SLC41A3 2.33 TOB1 2.58 COL5A2 2.49 ENST00000312289 2.40 WNK1 2.33 ENST00000382592 2.57 GAMT 2.49 TTYH3 2.40 RNF36 2.33 FAM117A 2.57 RSHL2 2.49 BC001335 2.40 BZW2 2.33 ENST00000374929 2.57 KDELC1 2.49 TMEM117 2.39 RBM5 2.33 MYO1A 2.57 C4orf24 2.48 MTMR3 2.39 AI613259 2.33 FASTKD5 2.57 DKFZp547C195 2.48 ZNF219 2.39 NUDT15 2.33 DNAJC16 2.57 FLJ10815 2.48 TP53TG3 2.39 RAB36 2.32 PTPRE 2.57 MEG3 2.48 ZCCHC10 2.39 SERTAD1 2.32 ZFR 2.57 A_24_P882309 2.48 SDK2 2.39 AQP2 2.32 TMEM142C 2.57 C9orf6 2.48 PTPLA 2.39 THC2290313 2.32 THC2433060 2.57 NPC2 2.48 ARF4 2.39 ABHD1 2.32 C11orf41 2.56 A_32_P71183 2.47 CYB561D1 2.39 TMEM32 2.32 RP11-217H1.1 2.56 AA610484 2.47 EAF1 2.39 SCYL3 2.32 ENST00000380357 2.56 SRP19 2.47 KIAA1450 2.39 ZNF697 2.32 LOC649294 2.56 A_24_P332911 2.47 ENST00000377093 2.39 MGC15523 2.31 TNFSF10 2.56 THC2310680 2.46 A_32_P31206 2.39 A_23_P329062 2.31 ATN1 2.56 THC2308298 2.46 THC2390306 2.39 PIK3R3 2.31 ZNF30 2.56 THC2373429 2.46 C6orf173 2.39 SLC35E3 2.31 SPR 2.56 TMEM159 2.46 A_32_P225768 2.38 A_24_P383802 2.31 ANUBL1 2.55 BC044628 2.46 THC2336549 2.38 MGC23909 2.31 PARD6G 2.54 BBS1 2.46 FUNDC1 2.38 THC2281731 2.31 KIAA1443 2.54 ORAOV1 2.46 LOC653857 2.38 HGF 2.31 UBE4B 2.54 WDR42A 2.45 LOC440396 2.38 RP4-742C19.3 2.31 CLCN3 2.54 FMO4 2.45 BAZ2B 2.38 PCGF3 2.31 PLEKHQ1 2.54 ELF4 2.44 SUMF1 2.38 SLC22A15 2.30 PPFIBP1 2.54 ATF6 2.44 AK123302 2.38 TMEM128 2.30 ITM2B 2.54 ENST00000382108 2.44 ENST00000262525 2.37 SLC27A1 2.30 ENST00000278949 2.54 ZCWPW1 2.44 ENST00000323198 2.37 POP1 2.29 AK092875 2.54 BE138567 2.44 CCL24 2.37 POLK 2.29 ALCAM 2.54 AL137705 2.44 ST5 2.37 BC037740 2.29 LOC151534 2.53 LOC643201 2.44 THC2281539 2.37 AI825645 2.29 AK098422 2.53 FOXC1 2.44 ZC3H12A 2.37 A_23_P28397 2.29 TMEM138 2.53 FZD6 2.44 A_23_P130639 2.36 A_24_P255836 2.29 KIAA0513 2.53 NISCH 2.44 ARL6IP5 2.36 DNAJA5 2.29 THC2427841 2.53 PEPD 2.44 RASGRF1 2.36 PNMA1 2.29 FLJ10781 2.53 AK056119 2.44 PFTK1 2.36 RAP2C 2.29 AL832758 2.53 ANKRD11 2.43 ZNF435 2.35 CENTA2 2.29 THC2314566 2.53 ATXN3 2.43 STK17A 2.35 PCMTD2 2.29 ABHD13 2.52 KIAA0888 2.43 BLCAP 2.35 ZFAND5 2.29 LPXN 2.52 LY6G5C 2.42 CB240827 2.35 A_23_P27460 2.29 AA743218 2.52 LOC51252 2.42 STXBP3 2.35 PARP3 2.29 A_24_P221105 2.52 C7 2.42 IMPAD1 2.34 SMUG1 2.28 EXOC7 2.52 MORN2 2.42 E2F7 2.34 AP1G1 2.28 STAT4 2.52 POMT1 2.42 CAPN10 2.34 GFM2 2.28 SLC34A1 2.52 CHURC1 2.42 TP53AP1 2.34 TM7SF2 2.28 A_24_P298604 2.51 GRRP1 2.42 FLJ32065 2.34 ARHGEF7 2.28 GM2A 2.51 VANGL1 2.41 STARD8 2.34 ENST00000379734 2.28 AK096685 2.50 DKFZP434A0131 2.41 CEP135 2.34 MFSD1 2.28

TABLE ST-1 CONT’D: UP-REGULATED GENES BY SAM SCORE Gene ID Score Gene ID Score Gene ID Score Gene ID Score AK123446 2.28 DFFB 2.20 HDDC2 2.14 KLHL3 2.09 TMEM34 2.28 TMEM57 2.20 C9orf89 2.14 DHFRL1 2.09 AK026368 2.27 LAX1 2.20 TSHB 2.14 KIFAP3 2.08 INHBC 2.27 SCC-112 2.20 MT1M 2.14 MT1A 2.08 PKD2 2.27 TMEM68 2.20 BX115105 2.13 AK092942 2.08 MAP2K6 2.27 BTBD10 2.20 KCNK6 2.13 OR5H1 2.08 HELB 2.27 IGBP1 2.20 ZNF407 2.13 TRSPAP1 2.08 RBMX 2.27 RP11-151A6.2 2.20 PPID 2.13 A_32_P177097 2.08 A_24_P75994 2.27 FAM98A 2.20 SETDB1 2.13 KIAA1279 2.08 ENC1 2.27 SOCS6 2.20 CASQ1 2.13 THC2336861 2.08 C1orf164 2.26 PNKD 2.20 TMEM118 2.13 CR626222 2.08 UHMK1 2.26 C1orf124 2.19 LRRC47 2.13 STX6 2.08 A_24_P670147 2.26 CDC42EP1 2.19 FBXO38 2.13 BC038355 2.08 HDAC6 2.26 THC2406147 2.19 LRRC25 2.13 ENST00000378953 2.08 ENST00000367545 2.25 AVPI1 2.19 FLJ36868 2.13 ACCN1 2.08 THC2248354 2.25 TOP2A 2.19 ZNF212 2.13 ERCC8 2.08 BG284526 2.25 ZNF33B 2.19 PCBP4 2.13 MYO6 2.08 LRRC56 2.25 MKL2 2.19 NOD9 2.12 THC2335955 2.08 ASAH2 2.25 ELN 2.19 UPF1 2.12 FAM43B 2.08 ZMYM5 2.25 ZMAT3 2.19 THC2401540 2.12 THC2343771 2.08 FRY 2.25 EIF2C1 2.18 THC2434618 2.12 BC037919 2.08 MRPL27 2.25 CEP164 2.18 MKNK1 2.12 FAM60A 2.08 MYH14 2.24 KIAA1632 2.18 C15orf23 2.12 MTERF 2.08 IL28RA 2.24 AK055641 2.18 BCOR 2.12 BCL2L14 2.07 FLJ22028 2.24 FECH 2.18 C6orf166 2.12 ANKRD6 2.07 BU561469 2.24 PHF16 2.18 KLHL9 2.12 AK124281 2.07 RP5-1022P6.2 2.24 TFR2 2.18 BX538248 2.12 TFG 2.07 BF446608 2.24 ZDHHC7 2.18 KIAA0556 2.12 ENST00000308118 2.07 RIPK3 2.24 THC2317830 2.17 THC2311946 2.11 AEBP2 2.07 BTG2 2.23 PPARD 2.17 EPAS1 2.11 SERPINB8 2.07 C5orf15 2.23 C14orf49 2.17 AY203961 2.11 C17orf59 2.06 THC2341060 2.23 MMP11 2.17 XPR1 2.11 FLJ34077 2.06 EID-3 2.23 ROD1 2.17 MLYCD 2.11 DLG2 2.06 AMPD3 2.23 ENST00000316634 2.17 KIF1B 2.11 BACE1 2.06 CD511705 2.23 THC2379531 2.17 RCBTB1 2.11 WDR19 2.06 LOC399744 2.23 AL365520 2.17 C18orf1 2.11 AJ420487 2.06 ASB7 2.23 DEPDC7 2.17 C16orf69 2.11 BC041926 2.06 MT1F 2.23 ATG4A 2.17 DDX58 2.10 RIC3 2.06 ENST00000374390 2.23 C18orf54 2.17 C1orf121 2.10 SEC14L1 2.06 CTSL2 2.23 CRLF3 2.16 KIAA1383 2.10 OSTM1 2.06 DKFZp762I137 2.22 LOC340061 2.16 LOC646564 2.10 RFX3 2.05 ZNF134 2.22 AK094167 2.16 ZNF135 2.10 RY1 2.05 CRHR2 2.22 AF116619 2.16 THC2284657 2.10 GNAZ 2.05 MYBPC2 2.22 FCRLM1 2.16 SCD5 2.10 MAP2K1IP1 2.05 THC2275950 2.22 ENST00000374860 2.16 SGTB 2.10 PNPLA4 2.05 A_24_P290114 2.22 OR7E24 2.16 KCTD21 2.10 CR590163 2.05 EIF5A2 2.22 ZNF641 2.15 CTNND1 2.10 REV3L 2.05 ABHD3 2.22 C11orf24 2.15 PCYT1A 2.09 KIAA1856 2.04 FBXW8 2.22 ALG10 2.15 MLLT7 2.09 THC2376817 2.04 NOTCH1 2.21 LMO7 2.15 C20orf23 2.09 METTL4 2.04 AK074614 2.21 AL049387 2.15 IGF2AS 2.09 LOC349114 2.04 GRINL1A 2.21 LOC550643 2.15 AK124515 2.09 ENST00000349637 2.04 ZNF498 2.21 BX537551 2.15 PHF8 2.09 RDHE2 2.04 A_24_P635355 2.21 ATP6V1A 2.15 CR617033 2.09 PLAT 2.04 DMXL1 2.21 FAM3C 2.15 C1orf25 2.09 AK055501 2.04 PEX11B 2.21 C14orf159 2.15 TRAK1 2.09 CGRRF1 2.04 UBE2D1 2.21 THC2281591 2.15 ENST00000322831 2.09 BE537483 2.04 TTC12 2.21 BG209623 2.14 PRKAG2 2.09 RAP1B 2.04 LOC201229 2.21 TRIM68 2.14 MYO9B 2.09 C9orf127 2.03

TABLE S-T1: WILD TYPE IR-RESPONSE DOWN-REGULATED GENES (3630)* Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) UNG -5.62 ENST00000318251 -3.78 ENST00000344771 -3.39 AKT1 -3.10 CR613972 -5.56 BG259069 -3.77 SNRPB -3.38 C19orf24 -3.10 PAQR4 -5.51 A_24_P195400 -3.77 MCM10 -3.38 PFKFB4 -3.10 C12orf34 -5.39 PYCR1 -3.77 TNFRSF12A -3.38 A_32_P172215 -3.09 CDC6 -5.37 BC031876 -3.75 PSMD11 -3.37 MLLT4 -3.09 ENST00000375256 -5.09 LRRC20 -3.74 CR609948 -3.36 ZFAND2B -3.09 DTL -4.96 GLOXD1 -3.72 MVK -3.35 CB853344 -3.09 ENO2 -4.87 AK054718 -3.72 SLC39A3 -3.35 KLHL22 -3.09 BC079833 -4.83 A_24_P50281 -3.72 RASSF1 -3.35 ELK1 -3.08 PPP5C -4.79 THC2407189 -3.71 THC2338942 -3.35 SLC43A1 -3.08 CCNE2 -4.76 OPN3 -3.70 HIST1H1A -3.35 TCOF1 -3.08 RBM21 -4.66 FKBP4 -3.70 RASSF7 -3.34 BRD2 -3.08 ETS2 -4.57 IMPDH1 -3.70 THC2405842 -3.34 AK057740 -3.07 KLHL23 -4.57 MRPL17 -3.68 POLD3 -3.33 ABL1 -3.07 SLC1A5 -4.48 SETD1A -3.68 THC2431161 -3.33 L07392 -3.07 SYMPK -4.48 BC064349 -3.67 WBP2 -3.33 GLT25D1 -3.07 C1QTNF5 -4.44 GRAP -3.67 THC2337923 -3.32 CENTG3 -3.07 MAZ -4.42 HNRPA0 -3.66 CBX8 -3.32 ENST00000297544 -3.06 ACY1L2 -4.42 EMD -3.64 SAMD1 -3.30 RAB1B -3.06 VAMP2 -4.41 MLLT7 -3.62 RFC4 -3.30 KHSRP -3.05 NCKIPSD -4.34 CR598370 -3.61 ENSA -3.30 RP1-93H18.5 -3.05 CHAF1B -4.34 EIF2S2 -3.60 SYNC1 -3.30 A_23_P251196 -3.05 FAM111B -4.30 STS -3.60 ATXN7L3 -3.30 TMEM106C -3.05 THC2357547 -4.28 GNG7 -3.59 MYOHD1 -3.29 FLJ22639 -3.04 MCM4 -4.21 UHRF1 -3.59 ING2 -3.29 LYPLA2 -3.04 TEAD4 -4.18 ADCK2 -3.58 A_32_P78488 -3.28 LRRC45 -3.04 FLJ39779 -4.18 AP1S1 -3.57 TCTE3 -3.27 GPR157 -3.03 FUT11 -4.17 POLD1 -3.57 ZNF687 -3.27 ACP5 -3.03 OXCT2 -4.15 FGFRL1 -3.56 FXR2 -3.27 LOC644063 -3.03 P2RY11 -4.11 DONSON -3.56 CNOT3 -3.26 A_24_P221485 -3.03 BC036361 -4.10 AA843546 -3.56 PAXIP1 -3.25 ARL6IP6 -3.02 NAALADL1 -4.09 ZNF395 -3.56 HDGF -3.25 GPR30 -3.02 CDCA7L -4.08 HCFC1R1 -3.54 SUV420H2 -3.25 A_24_P401124 -3.02 A_24_P375962 -4.05 STARD3 -3.52 DEK -3.24 SOX12 -3.02 THC2404842 -4.05 C3orf15 -3.52 THC2315164 -3.23 PES1 -3.02 RCC1 -4.04 DMC1 -3.51 C20orf59 -3.22 WNT10A -3.02 KLHL17 -4.03 AES -3.50 BE003490 -3.21 TSC22D2 -3.01 TMUB1 -4.02 A_24_P727215 -3.50 THC2309258 -3.21 FUZ -3.00 THC2343426 -4.01 C16orf55 -3.49 CD300C -3.21 A_24_P307046 -3.00 THC2320434 -4.00 CENTA1 -3.49 BF869497 -3.20 ZNHIT4 -2.99 THC2407148 -4.00 PTK2B -3.49 ZNF414 -3.20 FGFR1 -2.99 CDC25A -3.96 ASNA1 -3.49 THAP7 -3.19 A_32_P57002 -2.99 WDR76 -3.96 SYNJ2 -3.49 KLHDC3 -3.19 MSH2 -2.98 GPR146 -3.94 MGC13017 -3.48 TAGLN2 -3.18 PTGES2 -2.98 ATPBD3 -3.91 N4BP3 -3.48 CELSR3 -3.16 PFKFB3 -2.98 ARHGEF1 -3.90 TNKS1BP1 -3.47 MESP1 -3.16 ANP32D -2.98 FEN1 -3.90 ZMYND19 -3.46 FAM80A -3.16 IPO7 -2.97 BM932296 -3.87 HK2 -3.46 AF132203 -3.16 EXO1 -2.97 RPP25 -3.87 AP2A1 -3.45 TMEM121 -3.15 A_24_P594094 -2.96 ST6GALNAC6 -3.85 FSCN1 -3.44 VEGF -3.14 XRCC3 -2.96 GINS2 -3.85 SREBF1 -3.43 CLSPN -3.13 SET -2.96 BC022233 -3.85 GALNTL4 -3.43 BE816002 -3.13 A_32_P725839 -2.96 CDCA7 -3.85 LOC339344 -3.43 AKT1S1 -3.12 H1FX -2.96 MINK1 -3.85 BC009228 -3.41 NR1D2 -3.12 AGPAT2 -2.95 WWOX -3.84 NBL1 -3.40 SIT1 -3.12 ZNF23 -2.95 CR625518 -3.81 TIGD7 -3.39 E2F2 -3.10 C5orf25 -2.95

* ONLY GENES WITH SAM SCORE ≤ -2 ARE SHOWN

TABLE S-T1: WILD TYPE IR-RESPONSE DOWN-REGULATED GENES (3630)* Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) UBE2MP1 -2.95 ENST00000288548 -2.81 TOM1 -2.72 MGC16037 -2.63 RHOBTB2 -2.94 BM984383 -2.81 B3GAT3 -2.72 POLR3K -2.63 C9orf74 -2.94 ACSL3 -2.80 INSIG2 -2.72 CR612065 -2.62 BX094072 -2.94 GRPEL1 -2.80 LMBRD2 -2.72 A_24_P853302 -2.62 FJX1 -2.94 FLJ40542 -2.80 ZNF473 -2.72 RP3-402G11.5 -2.62 ENST00000332534 -2.94 LIMD1 -2.80 LGALS4 -2.72 A_24_P221601 -2.62 VLDLR -2.94 ENO1B -2.80 LY9 -2.72 CXorf24 -2.62 DDN -2.93 THC2363476 -2.79 ARL4D -2.71 PPP1R3E -2.62 CCDC71 -2.93 NFATC1 -2.79 THC2273298 -2.71 LOC388114 -2.62 NEDD4L -2.92 TGFB3 -2.79 LOC339123 -2.71 THC2339084 -2.62 ST3GAL1 -2.92 BAT1 -2.79 ADCK4 -2.71 LENG9 -2.62 OVOS2 -2.91 MBD3 -2.79 CENTB1 -2.71 VAT1 -2.62 THC2376568 -2.91 D21S2056E -2.79 TP73 -2.71 FLJ37453 -2.62 A_32_P133926 -2.90 KRT18 -2.78 SR-A1 -2.70 TOMM40 -2.62 A_24_P75708 -2.90 SPIB -2.78 A_24_P200962 -2.70 A_24_P834646 -2.61 LSM12 -2.89 BU618765 -2.78 CABLES1 -2.70 ENST00000378887 -2.61 BSPRY -2.89 C16orf74 -2.78 DTX3 -2.70 KIAA1715 -2.61 U2AF2 -2.88 PPME1 -2.78 GRPEL2 -2.70 B4GALT2 -2.61 P4HA1 -2.88 SLC4A7 -2.78 PIGW -2.70 CMKLR1 -2.61 NUPL1 -2.88 THC2339658 -2.78 PUM2 -2.70 THC2373490 -2.61 C9orf32 -2.87 ENST00000327299 -2.78 A_24_P92823 -2.70 FASN -2.61 PER1 -2.87 HIST1H1B -2.78 UAP1L1 -2.69 ACADS -2.61 THC2344809 -2.87 ENST00000383518 -2.78 NFKBIL1 -2.69 EIF5A -2.61 A_24_P771278 -2.87 GMPPB -2.78 CDC45L -2.69 AK058000 -2.61 GPR89A -2.87 LOC643431 -2.77 A_24_P714707 -2.69 CAMKK2 -2.61 THC2342255 -2.86 SUV39H1 -2.77 TRERF1 -2.69 ST3GAL4 -2.61 ENST00000312785 -2.86 TBC1D22B -2.77 NPR2 -2.68 HYAL2 -2.61 CHD8 -2.86 ARFRP1 -2.77 CDC7 -2.68 IPMK -2.61 ERO1L -2.86 ENST00000370290 -2.77 TAF6L -2.68 WDR79 -2.61 MGC2408 -2.86 ACTN4 -2.76 LOC389517 -2.68 SFRS16 -2.60 EPN1 -2.85 EML3 -2.76 RNPS1 -2.68 LOC442013 -2.60 PDAP1 -2.85 SMARCB1 -2.76 THC2406815 -2.67 EMID1 -2.60 RNF167 -2.85 LAT -2.76 AK123655 -2.67 DCPS -2.60 THC2374166 -2.85 PPAN -2.76 GCN5L2 -2.67 SIX5 -2.60 C10orf119 -2.85 A_32_P98854 -2.76 LRRC42 -2.67 CHDH -2.60 WHSC1 -2.85 ZRANB1 -2.75 MAD2L1 -2.67 TRIM65 -2.59 DVL1 -2.85 FBXL8 -2.75 KCTD20 -2.66 ENST00000357180 -2.59 A_24_P101742 -2.85 ABCB8 -2.75 DEPDC4 -2.66 JAKMIP1 -2.59 UHRF2 -2.84 CENPB -2.75 KAZALD1 -2.66 FKBP5 -2.59 ENST00000221462 -2.84 TCF20 -2.75 PHF17 -2.66 ENST00000292562 -2.59 ZADH2 -2.84 CAMK1D -2.75 GNB5 -2.65 GIPC1 -2.58 HIVEP2 -2.84 KIAA1285 -2.75 DAGLBETA -2.65 MCMDC1 -2.58 AA495894 -2.84 C19orf28 -2.75 TNPO1 -2.65 RAI1 -2.58 PDXP -2.84 A_24_P247616 -2.75 A_24_P810074 -2.65 HM13 -2.58 HDAC7A -2.84 COX19 -2.74 ZC3H10 -2.65 FAM120C -2.58 UNC5CL -2.84 C17orf69 -2.74 A_32_P108420 -2.65 PCGF1 -2.58 HMG20B -2.83 ZNF700 -2.74 KIAA1008 -2.64 BC030106 -2.58 BC004969 -2.83 SIPA1 -2.74 A4GALT -2.64 CCDC109A -2.58 SCAND1 -2.83 SCD -2.74 NP285481 -2.64 ANP32A -2.57 NOC2L -2.83 LOC388796 -2.74 BC047380 -2.64 FLJ31951 -2.57 ETV4 -2.83 CDT1 -2.73 ABTB1 -2.64 VEGFB -2.57 THC2268736 -2.83 FLJ14346 -2.73 THC2280176 -2.64 NR2F6 -2.57 TFDP3 -2.82 SLC8A3 -2.73 ZNF598 -2.64 FLJ10154 -2.57 LOC196394 -2.82 AW467174 -2.73 SERPINH1 -2.64 FUT7 -2.57 CCBL1 -2.82 NELF -2.73 RBM38 -2.64 TNF -2.57 FLNA -2.82 C3orf40 -2.73 CD248 -2.63 MGC13005 -2.57 RARG -2.82 KLC4 -2.73 ARHGDIA -2.63 THC2283170 -2.57 A_24_P927072 -2.81 FLJ25715 -2.73 ADM -2.63 SLC19A1 -2.57 CDK2 -2.81 C8orf55 -2.72 KCND2 -2.63 PELP1 -2.56

TABLE S-T1: WILD TYPE IR-RESPONSE DOWN-REGULATED GENES (3630)* Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) NUP205 -2.56 THC2309960 -2.49 RCC2 -2.44 BC036622 -2.39 U2AF1 -2.56 NIPSNAP1 -2.49 PSIP1 -2.44 ENST00000339367 -2.39 MMP25 -2.56 CHMP7 -2.49 GNAI2 -2.44 COL1A1 -2.39 FAM57A -2.55 AK023376 -2.49 KIAA0664 -2.44 DCC1 -2.38 TMEM44 -2.55 TRMU -2.49 RBM15B -2.44 ING1 -2.38 GTPBP1 -2.55 CHMP4B -2.49 C14orf122 -2.44 AK057198 -2.38 MPDU1 -2.55 FOXJ2 -2.49 MIDN -2.44 A_24_P203976 -2.38 CHRAC1 -2.55 P2RX5 -2.49 TMEM93 -2.44 AL522622 -2.38 TTC19 -2.55 CALML4 -2.49 ENST00000374537 -2.44 SOX7 -2.38 AK091337 -2.55 ANXA6 -2.49 STIP1 -2.44 A_32_P109495 -2.38 THC2377644 -2.55 HSF4 -2.49 HES4 -2.43 PRR7 -2.38 HIPK3 -2.54 ENST00000380635 -2.48 DDX55 -2.43 BTBD14B -2.38 PIGX -2.54 C14orf43 -2.48 BAT2 -2.43 76P -2.38 ZNF692 -2.54 ABHD11 -2.48 GOLGA2L1 -2.43 PTGES3 -2.38 PRKCSH -2.54 ISOC2 -2.48 TMEM86B -2.42 BQ072652 -2.38 AKAP8L -2.54 KDELC2 -2.48 TICAM1 -2.42 C14orf121 -2.37 C9orf41 -2.54 LOC442075 -2.48 HSD17B6 -2.42 MVD -2.37 UPP1 -2.53 VPRBP -2.48 GEMIN7 -2.42 ANKRD25 -2.37 DDX11 -2.53 ENST00000315293 -2.48 ENST00000359244 -2.42 ERN1 -2.36 THC2357608 -2.53 CD151 -2.48 TJP2 -2.42 AK095904 -2.36 THC2436745 -2.53 A_24_P101503 -2.48 RNASET2 -2.42 SPHK1 -2.36 IL16 -2.53 LY6E -2.48 HBEGF -2.42 ZNF318 -2.36 ST3GAL3 -2.53 THRAP3 -2.48 ZC3HAV1L -2.42 TMEM39B -2.36 KLF6 -2.52 RAC3 -2.48 CLN8 -2.42 TM9SF4 -2.36 PCQAP -2.52 RGS14 -2.48 BC045174 -2.42 HLA-G -2.36 BAG5 -2.52 DPP7 -2.48 ATG9A -2.42 DA180164 -2.36 RAD23A -2.52 ILK -2.48 UBQLN1 -2.42 NRF1 -2.36 THC2341944 -2.52 HIST3H2BB -2.47 CYB5R3 -2.42 AK125269 -2.36 A_24_P418498 -2.52 ENST00000311630 -2.47 RAB5C -2.41 PLCH2 -2.36 MAPK3 -2.52 ENST00000369731 -2.47 TPCN1 -2.41 DNAL4 -2.36 PDDC1 -2.52 AJ243810 -2.47 XPOT -2.41 RFC2 -2.36 D89937 -2.51 A_24_P625898 -2.47 DBP -2.41 DOCK6 -2.36 MTAP -2.51 ANKRD40 -2.47 CCDC28B -2.41 TCF3 -2.36 SEPN1 -2.51 C3orf21 -2.47 BQ071182 -2.41 NAGPA -2.36 WNK1 -2.51 A_32_P213948 -2.47 OMP -2.41 WBSCR1 -2.36 BC047032 -2.51 APLN -2.47 SNRPD1 -2.41 UNC93B1 -2.36 TRIT1 -2.51 UBL4A -2.47 JUNB -2.41 AF212044 -2.35 A_24_P92411 -2.51 NGLY1 -2.47 PIN1 -2.41 A_32_P173122 -2.35 SLC16A3 -2.51 AK128413 -2.47 ENST00000369158 -2.41 FPGS -2.35 MGC21830 -2.51 IQCC -2.46 AYTL2 -2.41 SEPW1 -2.35 PAK2 -2.51 BC092452 -2.46 ENST00000246083 -2.41 CRSP7 -2.35 GLI4 -2.51 ZNF525 -2.46 SCNN1B -2.41 ZNF263 -2.35 BQ008507 -2.51 A_24_P530690 -2.46 KDELR1 -2.41 KHK -2.35 ARRB2 -2.50 AF116678 -2.46 BF246504 -2.41 ALKBH5 -2.35 AF086071 -2.50 A_24_P470809 -2.46 PIGA -2.40 RAD51 -2.35 KIAA1919 -2.50 MGC40405 -2.45 C10orf95 -2.40 AK075052 -2.35 NT5M -2.50 A_24_P534290 -2.45 LOC150223 -2.40 THC2305336 -2.35 GMEB1 -2.50 C19orf22 -2.45 PCP2 -2.40 PRIM1 -2.35 DAPK3 -2.50 TMTC4 -2.45 KIAA0963 -2.40 PHLPP -2.35 NUDT9P1 -2.50 TNRC5 -2.45 A_24_P375870 -2.40 A_32_P76122 -2.35 GANAB -2.50 FASTK -2.45 CIRH1A -2.40 LOC152217 -2.35 LOC642711 -2.50 GRLF1 -2.45 IGF2BP1 -2.39 KSR1 -2.34 FAM40B -2.50 CD3EAP -2.45 C17orf76 -2.39 TRPM2 -2.34 PPP2R5B -2.50 ENST00000332107 -2.45 AK001998 -2.39 ATAD2 -2.34 CACNB3 -2.50 ENST00000308862 -2.45 ZNF514 -2.39 NOLC1 -2.34 A_24_P143653 -2.50 SIAH1 -2.45 PEX16 -2.39 DLST -2.34 A_24_P565908 -2.49 AW029229 -2.45 A_24_P170309 -2.39 ZNF160 -2.34 MRPS2 -2.49 LOC144097 -2.44 C6orf148 -2.39 HOMER1 -2.34 LRP2BP -2.49 UBA2 -2.44 ING5 -2.39 FEM1C -2.34 * ONLY GENES WITH SAM SCORE ≤ -2 ARE SHOWN

TABLE S-T1: WILD TYPE IR-RESPONSE DOWN-REGULATED GENES (3630)* Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) NRBP1 -2.34 SLC1A1 -2.28 CABIN1 -2.23 MRPL19 -2.18 PDK1 -2.34 BARD1 -2.28 OGG1 -2.23 THC2409354 -2.18 BC063441 -2.34 VGLL4 -2.28 C9orf91 -2.23 TNFRSF13B -2.18 LRP8 -2.34 ENST00000308269 -2.28 CHRNA5 -2.23 STT3A -2.18 A_24_P255865 -2.34 LOC152719 -2.28 MARK2 -2.22 TMEM142A -2.18 WDHD1 -2.34 ACOT4 -2.28 GLUD2 -2.22 ICA1 -2.18 HIST1H2AG -2.33 TMEM102 -2.28 SLFN13 -2.22 TPD52L2 -2.18 C1orf90 -2.33 FLJ12529 -2.28 ZMYND10 -2.22 ABHD12 -2.17 ENST00000378266 -2.33 ENST00000338956 -2.28 A_24_P187174 -2.22 AF086536 -2.17 STX7 -2.33 DA234975 -2.28 hCAP-H2 -2.22 MED9 -2.17 HES5 -2.33 APOA1BP -2.28 SUV39H2 -2.22 CDC2L6 -2.17 THC2369600 -2.33 ZNF84 -2.27 LOC643668 -2.22 PKN1 -2.17 DCLRE1B -2.33 GGA2 -2.27 HAGHL -2.22 MTHFD1L -2.17 dJ222E13.2 -2.33 PHF20 -2.27 TETRAN -2.22 RASA4 -2.17 A_24_P118721 -2.33 SCMH1 -2.27 CD300A -2.22 BC068045 -2.17 BX104493 -2.33 ZNF331 -2.27 SEPX1 -2.22 SMARCA3 -2.17 TYRO3 -2.33 MGC4562 -2.27 PRR6 -2.22 A_32_P46700 -2.17 SH2D5 -2.33 BU661610 -2.27 TUBA8 -2.22 AF131777 -2.17 CAMK2D -2.33 UBE2M -2.27 LOC653801 -2.21 USP5 -2.17 ITGB1BP2 -2.33 GPR35 -2.27 FZR1 -2.21 XTP3TPA -2.17 BX419129 -2.32 MKNK2 -2.27 PARVB -2.21 GINS3 -2.16 RAB43 -2.32 CSPP1 -2.26 EGR1 -2.21 TAF1A -2.16 BRF1 -2.32 BC040628 -2.26 KIAA2013 -2.21 C13orf25 -2.16 RPRC1 -2.32 09/01/2006 -2.26 THC2347074 -2.21 A_24_P410246 -2.16 NUDCD3 -2.32 DPP9 -2.26 AI971459 -2.21 ENST00000307366 -2.16 KCNAB2 -2.32 THC2340759 -2.26 ENST00000333392 -2.21 BU686376 -2.16 THC2316236 -2.32 MAFF -2.26 HSPB6 -2.21 LIMD2 -2.16 KCNQ5 -2.31 THC2428103 -2.26 GPIAP1 -2.21 SCAP1 -2.16 CLDN3 -2.31 C6orf111 -2.26 TMEM81 -2.21 ENST00000255667 -2.16 TSR2 -2.31 THC2440162 -2.26 CD22 -2.21 BRD9 -2.16 SLC2A1 -2.31 LAT2 -2.26 FADS1 -2.21 NOL5A -2.16 SNRP70 -2.31 A_24_P383751 -2.26 FKBP8 -2.20 USP37 -2.16 RAB11FIP1 -2.31 A_24_P213336 -2.26 A_24_P233560 -2.20 SLC25A11 -2.16 AXIN1 -2.31 GSTCD -2.26 THC2310236 -2.20 CDC91L1 -2.16 CENPQ -2.31 LOC554248 -2.25 MAPK14 -2.20 LOC115098 -2.16 SMN2 -2.30 UCN -2.25 LOC645158 -2.20 PNPLA6 -2.16 MAPK12 -2.30 FKBPL -2.25 CXorf34 -2.20 FAM109A -2.16 AF289611 -2.30 FLJ35220 -2.25 A_24_P16071 -2.20 STK24 -2.16 BHLHB3 -2.30 AW389914 -2.25 EPHX1 -2.20 NUP155 -2.16 FCER2 -2.30 JMJD1A -2.25 BCAS3 -2.20 C20orf20 -2.15 PCOLN3 -2.30 CBX1 -2.25 YARS -2.19 SH3GL1 -2.15 AF086011 -2.30 BRI3BP -2.25 C11orf75 -2.19 C19orf47 -2.15 SKP2 -2.30 C14orf24 -2.25 ITPKA -2.19 THC2409283 -2.15 BZRAP1 -2.30 MAD2L2 -2.24 A_24_P324214 -2.19 AGPAT7 -2.15 PYGB -2.30 AGPAT6 -2.24 KIAA0738 -2.19 SLC44A2 -2.15 ZNF428 -2.30 LOC200420 -2.24 RAD23B -2.19 FOXP4 -2.15 DKFZP761H1710 -2.30 CDCA4 -2.24 HPCA -2.19 RENBP -2.15 ANKZF1 -2.29 SCLY -2.24 SMO -2.19 CHKA -2.15 PQLC1 -2.29 CD2BP2 -2.24 ENST00000367740 -2.19 USF2 -2.15 BQ066852 -2.29 A_32_P136427 -2.24 XPO7 -2.19 ENST00000358396 -2.15 AK026811 -2.29 TBL2 -2.24 EPS8L1 -2.19 C22orf9 -2.15 KIAA1509 -2.29 PRPF19 -2.24 FIGNL1 -2.19 NANP -2.15 CLPTM1 -2.29 FBXO34 -2.24 DLG3 -2.19 SIAH2 -2.15 A_24_P401150 -2.29 A_24_P921801 -2.24 MGC34725 -2.18 BC053353 -2.15 SPAG4 -2.29 STC2 -2.23 PPP2R5A -2.18 FLOT1 -2.15 BC040064 -2.29 YOD1 -2.23 LOC389458 -2.18 A_32_P177300 -2.14 PPAT -2.29 HTF9C -2.23 FAM76B -2.18 THC2304728 -2.14 A_24_P170283 -2.28 TBXAS1 -2.23 A_32_P114268 -2.18 MAF1 -2.14 TRAIP -2.28 TRIM26 -2.23 NCL -2.18 THC2274524 -2.14 * ONLY GENES WITH SAM SCORE ≤ -2 ARE SHOWN

TABLE S-T1: WILD TYPE IR-RESPONSE DOWN-REGULATED GENES (3630)* Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) TSC22D4 -2.14 SENP1 -2.11 EFEMP2 -2.07 A_24_P502652 -2.04 CBX6 -2.14 ROGDI -2.11 ATAD1 -2.07 COCH -2.04 HSF2 -2.14 CHD3 -2.11 HDAC2 -2.07 ZDHHC17 -2.04 PICK1 -2.14 A_24_P651129 -2.11 ENST00000335459 -2.07 PRKAR1B -2.04 DNAJB2 -2.14 A_24_P273014 -2.11 BC033986 -2.07 C1orf63 -2.04 ENST00000376572 -2.14 EXOC6 -2.11 VPS26B -2.07 LOC63920 -2.04 A_24_P213321 -2.14 MYADM -2.10 RNASEH2A -2.07 FAM35A -2.03 TRIM14 -2.14 NOP17 -2.10 ENST00000367142 -2.07 A_32_P169142 -2.03 RBBP8 -2.14 THC2375588 -2.10 AK126245 -2.07 BX106493 -2.03 A_24_P626812 -2.14 ATP2A3 -2.10 THC2311196 -2.07 CHPF -2.03 MTMR12 -2.14 SMYD4 -2.10 KLF7 -2.07 SNRPA -2.03 A_24_P418786 -2.14 INPP5F -2.10 AK130705 -2.07 PPP1R13L -2.03 A_24_P289834 -2.14 ENST00000344035 -2.10 PIK3R5 -2.07 FADD -2.03 PLCXD1 -2.14 PTPN23 -2.10 CECR5 -2.06 SSBP4 -2.03 MPPE1 -2.13 LDLR -2.10 LOC375133 -2.06 PRG2 -2.03 NUFIP1 -2.13 APOE -2.10 KIAA0683 -2.06 THC2280849 -2.03 GEMIN4 -2.13 DEGS1 -2.10 ANKRD9 -2.06 AK055529 -2.03 LOC92312 -2.13 AK026497 -2.09 AA974271 -2.06 BRD3 -2.03 PHF2 -2.13 DNMT1 -2.09 FZD1 -2.06 SETDB1 -2.03 ITPR3 -2.13 SLC27A4 -2.09 JUN -2.06 CHCHD4 -2.03 A_32_P148407 -2.13 LOC440334 -2.09 ZBTB7C -2.06 UBQLN4 -2.03 C9orf75 -2.13 LOC286260 -2.09 FLJ22814 -2.06 ENST00000366699 -2.03 GMNN -2.13 CCDC106 -2.09 MGC45491 -2.06 EEF2K -2.03 A_32_P224926 -2.13 ENST00000359653 -2.09 A_24_P324250 -2.06 CHM -2.03 SURF5 -2.13 MATN4 -2.09 RTN4 -2.06 GPR135 -2.03 ZNF34 -2.13 CAMKK1 -2.09 MTERFD1 -2.06 RBM13 -2.03 RGC32 -2.13 GNL1 -2.09 CHST10 -2.06 ZNF136 -2.03 ZBED3 -2.13 COMTD1 -2.09 BCCIP -2.06 HNRPAB -2.02 USP21 -2.13 WIRE -2.09 A_24_P692600 -2.06 LMNB1 -2.02 C20orf27 -2.13 SFRS4 -2.09 BC040577 -2.06 STK35 -2.02 TMEM97 -2.13 C9orf93 -2.09 ZNF714 -2.06 LOC645249 -2.02 A_24_P612921 -2.13 PDE6G -2.09 PRPF3 -2.06 SCYL1 -2.02 PAQR7 -2.13 METTL1 -2.09 A_24_P358302 -2.06 LOC148898 -2.02 GBL -2.13 A_23_P208582 -2.09 SLC2A3 -2.05 THC2414638 -2.02 SELI -2.13 C14orf109 -2.09 POU3F1 -2.05 AK126814 -2.02 A_24_P418019 -2.13 HIST1H2BD -2.09 CD96 -2.05 AK123483 -2.02 ACSL1 -2.12 ERICH1 -2.09 IDI1 -2.05 THC2376224 -2.02 C10orf9 -2.12 THC2311599 -2.08 A_24_P891276 -2.05 FAM119A -2.02 PER2 -2.12 TMEM45A -2.08 PRKCBP1 -2.05 PWP2H -2.02 FH -2.12 AYP1p1 -2.08 YEATS4 -2.05 CTSC -2.02 EGLN1 -2.12 TAF5L -2.08 HNRPC -2.05 PLXNC1 -2.02 CX164944 -2.12 BF433725 -2.08 ANP32E -2.05 CR624880 -2.02 TRIB3 -2.12 SFRS1 -2.08 PRPS1 -2.05 GART -2.01 A_24_P349580 -2.12 RHEB -2.08 MLL3 -2.05 CTNNAL1 -2.01 POLE -2.12 GOLPH2 -2.08 FLJ39660 -2.05 AL833832 -2.01 WDR46 -2.12 A_23_P88554 -2.08 NR1H2 -2.05 YWHAG -2.01 PPP1R9B -2.12 RAD50 -2.08 A_24_P349756 -2.05 ENOSF1 -2.01 AFARP1 -2.12 T05215 -2.08 FADS2 -2.04 C19orf25 -2.01 IL17D -2.12 APH1A -2.08 COPA -2.04 TMEM153 -2.01 VKORC1 -2.12 ZFP36 -2.08 RAB33A -2.04 TMED9 -2.01 ENST00000270238 -2.12 A_24_P933940 -2.08 TAZ -2.04 XRCC2 -2.01 AK026675 -2.12 RALGDS -2.08 TNNC1 -2.04 AXUD1 -2.01 ADAM15 -2.12 PFKL -2.08 TERT -2.04 PSAP -2.01 BC067908 -2.12 CXorf9 -2.08 LOC283887 -2.04 AF086335 -2.01 KIAA1542 -2.12 SLC4A2 -2.08 A_24_P307424 -2.04 PRPS1L1 -2.01 ITGB1BP1 -2.11 BAT4 -2.08 SMC3 -2.04 A_24_P50294 -2.01 CCDC19 -2.11 EIF4EBP1 -2.07 EYA3 -2.04 ENST00000314984 -2.01 RNPEPL1 -2.11 BE467780 -2.07 SLC25A37 -2.04 RHBDL3 -2.00 CNP -2.11 BCAT2 -2.07 SLC43A3 -2.04 INSIG1 -2.00 * ONLY GENES WITH SAM SCORE ≤ -2 ARE SHOWN

100

TABLE S-T2: BRCA1 IR Classifier: Up-Regulated Features (466)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P112957' 0.026 'AK126405' 0.015 'BU540282' 0.030 'CR622844' 0.011 'A_23_P117943' 0.024 'AK128048' 0.042 'BU633383' 0.028 'CR626222' 0.022 'A_23_P9509' 0.009 'AL547361' 0.029 'BU733098' 0.049 'CREBL1' 0.030 'A_24_P161345' 0.012 'AL833005' 0.039 'BUD13' 0.038 'CSNK1G2' 0.015 'A_24_P16291' 0.012 'AL833005' 0.045 'BX352604' 0.012 'CSTF3' 0.032 'A_24_P229438' 0.035 'ANKRD11' 0.047 'BY798288' 0.036 'CUL5' 0.008 'A_24_P230457' 0.023 'ANKRD47' 0.010 'C10orf6' 0.024 'CXCL3' 0.034 'A_24_P289834' 0.035 'APBB1IP' 0.040 'C11orf56' 0.037 'CXorf15' 0.030 'A_24_P32930' 0.038 'APOL6' 0.028 'C12orf51' 0.035 'CYSLTR1' 0.033 'A_24_P409361' 0.021 'ARFGAP1' 0.017 'C14orf119' 0.040 'DAPK2' 0.047 'A_24_P417757' 0.040 'ARHGAP25' 0.010 'C14orf124' 0.028 'DB328061' 0.049 'A_24_P41781' 0.040 'ARIH2' 0.010 'C14orf147' 0.026 'DB380193' 0.030 'A_24_P600603' 0.026 'ARL17P1' 0.014 'C14orf2' 0.022 'DEFB4' 0.017 'A_24_P660797' 0.026 'ASCC3' 0.015 'C15orf27' 0.017 'DHRS7B' 0.009 'A_24_P67552' 0.048 'ASRGL1' 0.028 'C18orf17' 0.049 'DIAPH3' 0.021 'A_24_P837085' 0.040 'ASTN2' 0.000 'C19orf24' 0.023 'DLGAP4' 0.049 'A_24_P84268' 0.014 'ATXN7L2' 0.022 'C1orf162' 0.007 'DOCK3' 0.049 'A_24_P913620' 0.040 'AU185665' 0.040 'C1orf183' 0.047 'ELF2' 0.001 'A_24_P928250' 0.026 'AW168145' 0.004 'C1orf31' 0.024 'ELMO3' 0.020 'A_32_P10643' 0.017 'AW327568' 0.015 'C1orf54' 0.013 'ENST00000219169' 0.035 'A_32_P107797' 0.030 'AW804491' 0.016 'C20orf12' 0.039 'ENST00000242224' 0.021 'A_32_P109645' 0.025 'AW946823' 0.044 'C20orf6' 0.041 'ENST00000265271' 0.036 'A_32_P147603' 0.002 'AY170823' 0.027 'C21orf119' 0.008 'ENST00000297812' 0.041 'A_32_P187050' 0.016 'AY562498' 0.034 'C22orf15' 0.046 'ENST00000298453' 0.048 'A_32_P220897' 0.019 'AY672103' 0.042 'C3orf19' 0.031 'ENST00000307106' 0.010 'A_32_P42149' 0.020 'AY927536' 0.043 'C3orf54' 0.040 'ENST00000307366' 0.033 'A_32_P51714' 0.022 'AYP1p1' 0.025 'C5orf24' 0.026 'ENST00000327423' 0.021 'A_32_P57002' 0.020 'B3GALT6' 0.033 'C5orf4' 0.000 'ENST00000329309' 0.019 'A_32_P64025' 0.025 'BC001764' 0.013 'C6orf148' 0.018 'ENST00000329627' 0.012 'A_32_P83795' 0.026 'BC007968' 0.015 'C6orf52' 0.024 'ENST00000333392' 0.014 'A_32_P8971' 0.003 'BC011779' 0.028 'C9orf102' 0.049 'ENST00000342294' 0.037 'A_32_P91156' 0.019 'BC011998' 0.035 'C9orf3' 0.041 'ENST00000355629' 0.046 'A_32_P95502' 0.045 'BC015588' 0.038 'CA420826' 0.040 'ENST00000357778' 0.049 'A_32_P9924' 0.039 'BC019044' 0.012 'CA429641' 0.020 'ENST00000361227' 0.012 'AA427821' 0.033 'BC040051' 0.023 'CABYR' 0.008 'ENST00000361681' 0.020 'AA594808' 0.006 'BC046172' 0.025 'CALML3' 0.042 'ENST00000366699' 0.026 'AA725389' 0.021 'BC053363' 0.048 'CARKL' 0.029 'ENST00000372602' 0.046 'AA725860' 0.041 'BC054888' 0.008 'CASC4' 0.023 'ENST00000372866' 0.007 'AA780798' 0.031 'BCL2A1' 0.032 'CASQ1' 0.021 'ENST00000373268' 0.048 'AA971667' 0.044 'BE551270' 0.041 'CB993898' 0.009 'ENST00000376068' 0.018 'AA991488' 0.009 'BE564275' 0.016 'CBX1' 0.014 'ENST00000378536' 0.040 'ACTR2' 0.018 'BE646426' 0.021 'CCL17' 0.007 'EP300' 0.031 'ADAMTS12' 0.016 'BE769489' 0.017 'CCNK' 0.005 'EPC2' 0.038 'AF070529' 0.018 'BEX2' 0.024 'CCR10' 0.011 'EPM2A' 0.048 'AF132973' 0.023 'BF378976' 0.041 'CD671824' 0.045 'EPN3' 0.020 'AF271776' 0.030 'BF718543' 0.016 'CDADC1' 0.035 'ERICH1' 0.040 'AF289590' 0.019 'BF827074' 0.040 'CDC2L1' 0.009 'FAM111B' 0.014 'AGMAT' 0.019 'BF870807' 0.033 'CDC42' 0.009 'FAM121B' 0.045 'AHCTF1' 0.028 'BG058000' 0.014 'CDK9' 0.044 'FAM98A' 0.015 'AI079353' 0.001 'BG205572' 0.033 'CEACAM1' 0.004 'FANCF' 0.033 'AI143116' 0.049 'BG775850' 0.016 'CF528315' 0.030 'FBXO31' 0.049 'AI225215' 0.046 'BI262095' 0.026 'CHCHD5' 0.020 'FCRLM2' 0.013 'AI608782' 0.004 'BM474343' 0.044 'CKLF' 0.022 'FGFR1' 0.045 'AI818152' 0.032 'BM677531' 0.012 'CMTM4' 0.007 'FKBP1B' 0.047 'AIF1' 0.023 'BM702886' 0.039 'COL10A1' 0.047 'FLJ12949' 0.017 'AJ243810' 0.044 'BM955917' 0.047 'COMMD5' 0.025 'FLJ13305' 0.037 'AK023660' 0.037 'BM982926' 0.041 'CORO6' 0.038 'FLJ14346' 0.022 'AK024035' 0.023 'BOLA1' 0.033 'CR597807' 0.032 'FLJ21127' 0.042 'AK057740' 0.017 'BQ000122' 0.033 'CR612065' 0.008 'FLJ39779' 0.039 *Features "called" more than once only represented once in table, total # called features = 1110 (up & down)

101

TABLE S-T2: BRCA1 IR Classifier: Up-Regulated Features (466)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'FMR1' 0.048 'LSM12' 0.045 'RANGNRF' 0.048 'THC2376140' 0.020 'FUNDC2' 0.046 'MAP3K10' 0.028 'RASSF3' 0.045 'THC2378276' 0.041 'GABARAPL2' 0.047 'MAP3K12' 0.045 'RCP9' 0.026 'THC2395355' 0.044 'GALNT11' 0.025 'MCFD2' 0.026 'REEP2' 0.004 'THC2406815' 0.011 'GHRL' 0.048 'MED9' 0.048 'RGS12' 0.003 'THC2411070' 0.032 'GIMAP5' 0.043 'MGC13379' 0.047 'RP11-535K18.3' 0.030 'THC2425829' 0.038 'GIMAP6' 0.049 'MGC16121' 0.037 'RP9' 0.043 'THC2425852' 0.023 'GMEB1' 0.034 'MGC16597' 0.048 'RPE' 0.027 'THC2436485' 0.033 'GPC2' 0.020 'MGC21830' 0.012 'RPS9' 0.032 'THC2437618' 0.016 'GPR84' 0.022 'MGC33556' 0.047 'RSN' 0.022 'THC2440787' 0.032 'GTF2A1' 0.029 'MGC4093' 0.037 'S81524' 0.028 'THC2441239' 0.045 'HAVCR2' 0.047 'MGC52282' 0.044 'SAC3D1' 0.021 'THC2442550' 0.003 'HB-1' 0.025 'MICAL-L2' 0.006 'SAPS2' 0.026 'THRAP2' 0.031 'HBEGF' 0.037 'MOCOS' 0.031 'SART2' 0.050 'TIGD6' 0.014 'HEAB' 0.041 'MRPL17' 0.038 'SEPW1' 0.015 'TIMP1' 0.047 'HEMK1' 0.038 'MRPS16' 0.041 'SERTAD1' 0.046 'TMEFF2' 0.046 'HINT2' 0.024 'MTF1' 0.024 'SIGIRR' 0.030 'TMEM103' 0.034 'HIST1H1A' 0.049 'MTHFS' 0.031 'SIRPA' 0.046 'TMEM107' 0.048 'HIST1H1B' 0.019 'MXD1' 0.036 'SLC12A4' 0.043 'TMEM111' 0.045 'HIST1H2AB' 0.029 'MYO1A' 0.020 'SLC12A9' 0.048 'TMEM132A' 0.043 'HIST1H4B' 0.001 'MYST4' 0.023 'SLC31A2' 0.037 'TMEM136' 0.019 'HIST1H4D' 0.002 'N75427' 0.009 'SLC35F3' 0.038 'TMEM53' 0.038 'HIST1H4E' 0.000 'N91552' 0.041 'SLC6A4' 0.038 'TMEM58' 0.012 'HIST1H4J' 0.037 'NCR3' 0.049 'SMG6' 0.024 'TNFSF14' 0.026 'HIST1H4K' 0.028 'NENF' 0.004 'SNAPAP' 0.038 'TNNI2' 0.016 'HIST2H2AB' 0.044 'NEXN' 0.046 'SP140' 0.034 'TP73L' 0.025 'HIST2H2AC' 0.008 'NGFRAP1' 0.050 'SPFH2' 0.019 'TPST1' 0.021 'HIST3H2BB' 0.039 'NOL8' 0.030 'SPN' 0.014 'TRO' 0.030 'HLA-DOB' 0.032 'NUDT16' 0.022 'SSX2' 0.025 'TSPAN10' 0.015 'HNRPA3' 0.020 'OR7E156P' 0.037 'STOML1' 0.039 'TUSC2' 0.020 'HUWE1' 0.037 'P117' 0.031 'STX2' 0.029 'U12206' 0.020 'HYAL3' 0.044 'PANK2' 0.005 'T12588' 0.027 'UBA52' 0.041 'ID3' 0.027 'PCNT' 0.020 'TAF2' 0.042 'UBE2Q1' 0.050 'IMP3' 0.011 'PDCD7' 0.047 'TAF5L' 0.030 'UGT1A6' 0.018 'INCENP' 0.045 'PDE4C' 0.032 'TAL1' 0.048 'UROD' 0.026 'INTS2' 0.044 'PHF2' 0.042 'TANC1' 0.040 'USP20' 0.030 'ISOC2' 0.002 'PHF6' 0.047 'TCF25' 0.032 'USP51' 0.003 'ITSN2' 0.029 'PHIP' 0.041 'TDG' 0.006 'VPREB3' 0.039 'JAGN1' 0.043 'PHKG2' 0.036 'THC2265453' 0.044 'WASF2' 0.024 'JMJD1C' 0.043 'PIGL' 0.035 'THC2265980' 0.039 'WDR8' 0.034 'JUNB' 0.048 'PNPLA4' 0.011 'THC2268341' 0.026 'WDR90' 0.029 'KCTD9' 0.012 'POLD2' 0.026 'THC2268405' 0.025 'X62691' 0.014 'KIF21A' 0.008 'POLR2I' 0.013 'THC2269172' 0.041 'YTHDC2' 0.039 'KIF9' 0.033 'POM121' 0.043 'THC2270072' 0.031 'ZAP70' 0.040 'LARS' 0.032 'POP1' 0.017 'THC2271549' 0.034 'ZC3H6' 0.041 'LOC145853' 0.047 'PPM1K' 0.045 'THC2271582' 0.002 'ZCCHC2' 0.019 'LOC147650' 0.018 'PRKAB2' 0.028 'THC2274051' 0.002 'ZFYVE19' 0.030 'LOC203547' 0.045 'PRO2900' 0.015 'THC2274074' 0.008 'ZFYVE9' 0.023 'LOC205251' 0.009 'PRR4' 0.039 'THC2277571' 0.027 'ZNF3' 0.013 'LOC283481' 0.022 'PRRG4' 0.010 'THC2281903' 0.048 'ZNF576' 0.046 'LOC283887' 0.020 'PSCD2L' 0.042 'THC2307560' 0.036 'ZNF644' 0.049 'LOC388965' 0.029 'PSRC2' 0.044 'THC2311248' 0.015 'ZNF688' 0.047 'LOC440288' 0.006 'PTPRE' 0.046 'THC2312748' 0.046 'LOC440455' 0.048 'PURA' 0.027 'THC2335347' 0.031 'LOC441212' 0.033 'PXMP4' 0.011 'THC2337454' 0.030 'LOC619208' 0.020 'RAB5C' 0.031 'THC2345392' 0.012 'LOC644053' 0.045 'RABIF' 0.037 'THC2361211' 0.027 'LRDD' 0.020 'RAGE' 0.039 'THC2366887' 0.039

102

TABLE S-T2: BRCA1 IR Classifier: Down-Regulated Features (544)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P151376' 0.007 'ARHGAP15' 0.010 'CD2AP' 0.047 'EBF' 0.036 'A_23_P399579' 0.036 'ARHGEF12' 0.041 'CD46' 0.032 'EDD1' 0.007 'A_23_P96017' 0.049 'ARID1A' 0.018 'CDC37L1' 0.039 'EDG1' 0.043 'A_24_P144314' 0.003 'ARIH1' 0.046 'CDC42SE2' 0.040 'EIF2S3' 0.010 'A_24_P178333' 0.050 'ARMCX3' 0.046 'CDC73' 0.040 'EIF3S6' 0.030 'A_24_P178643' 0.046 'ASCIZ' 0.028 'CDK2' 0.034 'EIF4A2' 0.034 'A_24_P255473' 0.025 'ATAD1' 0.045 'CDKN1B' 0.014 'EIF4E' 0.010 'A_24_P290257' 0.019 'ATF1' 0.029 'CDV3' 0.017 'EIF5B' 0.042 'A_24_P298744' 0.039 'ATG4C' 0.025 'CENPL' 0.033 'ELMO2' 0.024 'A_24_P332532' 0.025 'ATP6V1C1' 0.005 'CENTB2' 0.035 'ENST00000216214' 0.015 'A_24_P332911' 0.010 'ATP6V1D' 0.040 'CFDP1' 0.024 'ENST00000246024' 0.025 'A_24_P341593' 0.031 'ATPAF1' 0.038 'CGGBP1' 0.016 'ENST00000265450' 0.011 'A_24_P349648' 0.025 'ATRX' 0.024 'CGI-09' 0.049 'ENST00000269142' 0.045 'A_24_P358302' 0.045 'ATXN10' 0.044 'CHCHD3' 0.020 'ENST00000276938' 0.033 'A_24_P358868' 0.025 'AV698092' 0.048 'CHMP2B' 0.049 'ENST00000282964' 0.024 'A_24_P417751' 0.043 'AW291149' 0.041 'CKAP2' 0.025 'ENST00000290786' 0.041 'A_24_P455100' 0.036 'AW663344' 0.034 'CLASP2' 0.036 'ENST00000303246' 0.010 'A_24_P464798' 0.015 'AZI2' 0.037 'CLCC1' 0.034 'ENST00000323501' 0.027 'A_24_P474188' 0.027 'B4GALNT1' 0.020 'CLIC4' 0.038 'ENST00000325863' 0.024 'A_24_P50390' 0.023 'BACH1' 0.014 'CLK1' 0.036 'ENST00000334464' 0.014 'A_24_P670147' 0.048 'BAG5' 0.026 'CLPX' 0.004 'ENST00000340855' 0.027 'A_24_P683734' 0.027 'BBS7' 0.013 'CMAS' 0.037 'ENST00000341262' 0.045 'A_24_P713893' 0.029 'BC038355' 0.047 'CNN3' 0.027 'ENST00000341591' 0.040 'A_24_P75688' 0.036 'BC047111' 0.037 'CNOT1' 0.007 'ENST00000355854' 0.019 'A_24_P846755' 0.041 'BF509482' 0.042 'CNOT4' 0.026 'ENST00000358951' 0.041 'A_24_P84873' 0.049 'BF965065' 0.020 'CNOT6' 0.047 'ENST00000367740' 0.022 'A_24_P859431' 0.040 'BG108194' 0.032 'COG3' 0.033 'ENST00000369308' 0.044 'A_24_P882666' 0.012 'BMS1L' 0.012 'CORO1C' 0.021 'ENST00000369453' 0.008 'A_24_P900721' 0.028 'BP290435' 0.043 'CPNE3' 0.023 'ENST00000369705' 0.025 'A_24_P938006' 0.003 'BP75' 0.032 'CPNE8' 0.012 'ENST00000373954' 0.001 'A_32_P6274' 0.042 'BRD7' 0.014 'CPOX' 0.033 'ENST00000374285' 0.040 'AA585242' 0.006 'BRDG1' 0.027 'CR591764' 0.020 'ENST00000376573' 0.009 'ABCD3' 0.023 'BRWD1' 0.035 'CR610374' 0.029 'ENST00000378624' 0.028 'ABHD2' 0.036 'BU173515' 0.023 'CR622308' 0.022 'ENST00000379156' 0.017 'ABI1' 0.008 'C10orf46' 0.037 'CR623684' 0.024 'ENST00000379287' 0.030 'ACADSB' 0.024 'C11orf46' 0.025 'CREBBP' 0.034 'ENST00000379870' 0.023 'ACBD3' 0.048 'C14orf104' 0.036 'CRLF3' 0.003 'ENST00000379969' 0.040 'ACOT9' 0.030 'C16orf63' 0.037 'CSDE1' 0.004 'ENST00000381298' 0.046 'ACTR8' 0.043 'C17orf27' 0.045 'CSNK1G3' 0.016 'ENST00000382108' 0.023 'ADAM33' 0.014 'C17orf80' 0.049 'CSNK2A1' 0.032 'ENST00000382990' 0.005 'AF088033' 0.035 'C1orf149' 0.004 'CTDSPL2' 0.019 'EPB41L2' 0.043 'AF445027' 0.017 'C1orf80' 0.012 'CTNNB1' 0.038 'ERCC8' 0.040 'AHCTF1' 0.017 'C1QDC1' 0.007 'CTSO' 0.023 'ERO1L' 0.042 'AHI1' 0.044 'C20orf45' 0.016 'CTSS' 0.004 'ESR2' 0.023 'AK021546' 0.027 'C3orf34' 0.035 'CYP51A1' 0.037 'EXOD1' 0.022 'AK024382' 0.017 'C3orf58' 0.017 'D80006' 0.012 'FAM18B2' 0.032 'AK057591' 0.037 'C5orf15' 0.014 'DCAKD' 0.043 'FAM21C' 0.035 'AK057981' 0.025 'C6orf151' 0.005 'DDX50' 0.032 'FAM29A' 0.032 'AK074960' 0.011 'C6orf167' 0.028 'DENND4A' 0.019 'FAM33A' 0.021 'AK092090' 0.013 'C6orf62' 0.035 'DENR' 0.006 'FAM3C' 0.004 'AK092888' 0.027 'C6orf68' 0.040 'DHRS7' 0.006 'FAM76A' 0.038 'AL117599' 0.027 'CAPN7' 0.045 'DKFZp547C195' 0.040 'FBXO3' 0.043 'AL117621' 0.042 'CAPZA1' 0.023 'DKFZP564C152' 0.046 'FBXW8' 0.013 'ANKHD1' 0.029 'CAV1' 0.003 'DKFZP564O0523' 0.042 'FLJ12716' 0.024 'ANKHD1' 0.035 'CCDC100' 0.036 'DKFZP779L1068' 0.018 'FLJ21908' 0.006 'ANKRD28' 0.032 'CCDC117' 0.032 'DLG1' 0.016 'FLJ30596' 0.011 'APC' 0.019 'CCDC55' 0.014 'DNAJA2' 0.020 'FLJ38663' 0.048 'APOL2' 0.037 'CCNG1' 0.022 'DNAJB4' 0.016 'FLJ43934' 0.027 'ARCN1' 0.014 'CD164' 0.037 'DYNC1LI2' 0.046 'FLJ44894' 0.036

103

TABLE S-T2: BRCA1 IR Classifier: Down-Regulated Features (544)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'FUBP3' 0.026 'LRRC47' 0.029 'PLDN' 0.029 'SAPS3' 0.050 'FUSIP1' 0.045 'LRRFIP1' 0.023 'PLXNC1' 0.049 'SAT' 0.011 'GALNACT-2' 0.020 'LRRIQ2' 0.038 'PMS1' 0.044 'SBNO1' 0.011 'GBAS' 0.015 'LSM14A' 0.020 'PPARA' 0.020 'SC4MOL' 0.009 'GCA' 0.045 'LUC7L2' 0.041 'PPID' 0.013 'SCAP2' 0.019 'GMPS' 0.005 'LY75' 0.031 'PPP1CB' 0.029 'SERBP1' 0.037 'GNPTAB' 0.049 'MALT1' 0.037 'PPP1CC' 0.025 'SERPINB9' 0.029 'GOLGA7' 0.001 'MAP2K1IP1' 0.034 'PPP1R2' 0.014 'SFRS1' 0.023 'GOLPH3' 0.014 'MAP3K7IP3' 0.024 'PPP2CA' 0.031 'SH3BGRL' 0.040 'GPIAP1' 0.037 'MAP4K5' 0.044 'PPP2R2A' 0.011 'SHOC2' 0.046 'GPR174' 0.009 'MAPK9' 0.019 'PPP3CA' 0.045 'SHPRH' 0.005 'GPSM2' 0.048 'Mar-05' 0.033 'PPP3CB' 0.040 'SIPA1L1' 0.043 'GSTM4' 0.043 'Mar-07' 0.010 'PRDX3' 0.045 'SKIV2L2' 0.041 'GTF2I' 0.035 'MAT2B' 0.015 'PRMT3' 0.017 'SLA/LP' 0.041 'GYG1' 0.035 'MBP' 0.041 'PRPS2' 0.027 'SLAMF6' 0.013 'HDDC2' 0.027 'MCCC2' 0.026 'PRPSAP2' 0.050 'SLC30A5' 0.016 'HERC4' 0.039 'ME2' 0.028 'PSIP1' 0.035 'SLC35A3' 0.002 'HEXB' 0.047 'METAP1' 0.030 'PTD004' 0.045 'SLC35D2' 0.019 'HIBADH' 0.012 'MIB1' 0.033 'PTER' 0.022 'SLC35F2' 0.037 'HIF1A' 0.050 'MIER1' 0.006 'PTPN11' 0.050 'SLC39A11' 0.030 'HMGN4' 0.012 'MKRN2' 0.029 'PUM1' 0.002 'SLC4A7' 0.024 'HNRPD' 0.039 'MOAP1' 0.016 'PUM2' 0.034 'SLC8A3' 0.046 'HOOK3' 0.031 'MOBK1B' 0.011 'QRSL1' 0.020 'SMAD2' 0.013 'HPS3' 0.036 'MORF4' 0.028 'QSCN6L1' 0.015 'SNAP23' 0.014 'IDH3A' 0.040 'MORF4L1' 0.017 'R3HDM2' 0.006 'SNX3' 0.029 'IDI1' 0.045 'MOSPD2' 0.015 'RAB21' 0.007 'SOS1' 0.042 'IFIH1' 0.033 'MRPL44' 0.027 'RAB40B' 0.004 'SOX4' 0.045 'IFNAR1' 0.032 'MTAP' 0.044 'RAB5A' 0.017 'SPAST' 0.027 'INADL' 0.032 'MTPN' 0.025 'RAB6C' 0.039 'SPEN' 0.032 'ING1' 0.021 'MTRF1L' 0.048 'RAB7L1' 0.026 'SPIN' 0.011 'INPP5A' 0.037 'NARG1' 0.026 'RABEP1' 0.035 'SRPK2' 0.004 'IPO8' 0.036 'NARS' 0.045 'RAD17' 0.032 'SSFA2' 0.020 'IQWD1' 0.024 'NAT12' 0.039 'RAD23B' 0.032 'ST13' 0.012 'IREB2' 0.031 'NBR1' 0.040 'RAD50' 0.033 'STAG2' 0.043 'ITPR1' 0.009 'NDUFA10' 0.028 'RALBP1' 0.047 'STARD13' 0.021 'JAK2' 0.018 'NEDD9' 0.039 'RANBP9' 0.006 'STAT1' 0.023 'KDELR2' 0.030 'NIPA1' 0.012 'RAP2C' 0.032 'STRN3' 0.035 'KIAA1411' 0.019 'NIPSNAP3B' 0.021 'RASA1' 0.049 'STYX' 0.046 'KIAA1432' 0.006 'NMD3' 0.032 'RASSF2' 0.028 'SUCLG2' 0.011 'KIAA1450' 0.030 'NMT2' 0.024 'RASSF5' 0.024 'SWAP70' 0.011 'KIF2' 0.015 'NPTN' 0.037 'RB1' 0.031 'SYNCRIP' 0.035 'KIF5B' 0.045 'NT5C2' 0.017 'RBL2' 0.045 'SYNE2' 0.038 'KIFAP3' 0.041 'NUDT15' 0.007 'RBM3' 0.049 'TADA1L' 0.017 'KRAS' 0.017 'NUDT4' 0.042 'RFFL' 0.003 'TAF11' 0.020 'LACTB' 0.040 'OPA1' 0.040 'RHEB' 0.030 'TAF1L' 0.026 'LNX2' 0.012 'OSBPL8' 0.038 'RHOBTB3' 0.028 'TBC1D24' 0.049 'LOC129285' 0.035 'OTUD4' 0.036 'RHOQ' 0.041 'TCEA1' 0.045 'LOC152024' 0.028 'P18SRP' 0.032 'RHOT1' 0.020 'TCP1' 0.041 'LOC389831' 0.031 'PAIP1' 0.011 'RHPN2' 0.042 'TDRD3' 0.031 'LOC392506' 0.024 'PBEF1' 0.032 'RIOK3' 0.049 'TERF2IP' 0.020 'LOC402643' 0.046 'PCYOX1' 0.025 'RNF111' 0.026 'TFCP2' 0.046 'LOC440577' 0.034 'PDCD4' 0.000 'RNF149' 0.028 'TFG' 0.021 'LOC441294' 0.038 'PDHX' 0.031 'RNF36' 0.031 'THAP9' 0.028 'LOC51136' 0.024 'PDIK1L' 0.024 'RNGTT' 0.006 'THC2250386' 0.024 'LOC57149' 0.033 'PELI1' 0.010 'RP11-82K18.3' 0.037 'THC2278663' 0.015 'LOC648674' 0.015 'PGRMC2' 0.012 'RP2' 0.023 'THC2279352' 0.032 'LOC653458' 0.017 'PICALM' 0.025 'RSL1D1' 0.008 'THC2279548' 0.018 'LOC90355' 0.039 'PITPNB' 0.020 'RSU1' 0.014 'THC2281165' 0.007 'LOC90624' 0.010 'PLD1' 0.015 'SAMSN1' 0.004 'THC2313495' 0.011

104

TABLE S-T2: BRCA1 IR Classifier: Down-Regulated Features Cont'd (544)* Gene Gene IDs p-value IDs p-value 'THC2397609' 0.047815484 'USP12' 0.033193788 'THC2400533' 0.01275521 'USP14' 0.031046614 'THC2404072' 0.045270598 'USP25' 0.033281716 'THUMPD3' 0.030794735 'USP28' 0.044010869 'TICAM2' 0.003870305 'USP33' 0.012580603 'TIPRL' 0.041130456 'UTP18' 0.030229424 'TLK1' 0.000723309 'VDAC1' 0.035611184 'TLOC1' 0.007258043 'VPS13B' 0.032000747 'TLR10' 0.042338555 'VPS24' 0.035380708 'TMED7' 0.022798207 'VPS26A' 0.041378858 'TMEM135' 0.012884692 'VPS41' 0.037113236 'TMEM159' 0.043415348 'VPS4B' 0.037215295 'TMEM23' 0.046736286 'WDFY1' 0.018915109 'TMEM30A' 0.025312393 'WDR20' 0.035830561 'TNFSF5IP1' 0.032671142 'XPOT' 0.042543518 'TNKS' 0.007659245 'YEATS4' 0.012846442 'TOR1AIP1' 0.023717661 'YES1' 0.026995921 'TPM3' 0.043371685 'YWHAZ' 0.006724257 'TRIB1' 0.03893363 'ZBTB11' 0.016891178 'TRIM22' 0.023213509 'ZBTB26' 0.029949239 'TRIM4' 0.030401687 'ZFAND1' 0.006512656 'TRIM44' 0.026490954 'ZFYVE16' 0.023957646 'TRIM5' 0.029663752 'ZMYM1' 0.023701903 'TROVE2' 0.001095428 'ZMYM6' 0.043121108 'TTF2' 0.041240687 'ZNF267' 0.017929088 'TUBE1' 0.02915967 'ZNF302' 0.011343855 'TUBGCP3' 0.038706914 'ZNF397' 0.037704568 'TUG1' 0.042382688 'ZNF429' 0.039376536 'UAP1' 0.04255059 'ZNF451' 0.032952718 'UBE1C' 0.013712091 'ZNF468' 0.0417563 'UBE2D1' 0.039331696 'ZNF529' 0.01672718 'UBE2H' 0.030657554 'ZNF545' 0.016158778 'UBE3A' 0.035272932 'ZNF566' 0.029506494 'UBXD4' 0.043602636 'ZNF616' 0.022510332 'UFM1' 0.008994037 'ZNF639' 0.022287147 'UNC84A' 0.024540967 'ZNF92' 0.028345819

105

TABLE S-T3: BRCA2 IR Classifier: Up-Regulated Features (214)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P147404' 0.029 'C7orf30' 0.020 'HRASLS3' 0.003 'SLC6A4' 0.013 'A_24_P161068' 0.037 'CCDC26' 0.015 'ICA1' 0.048 'SOCS2' 0.019 'A_24_P170203' 0.031 'CD518214' 0.013 'IL23A' 0.034 'SORBS3' 0.020 'A_24_P178167' 0.033 'CDC42' 0.011 'KIAA1109' 0.037 'SOX4' 0.025 'A_24_P204474' 0.027 'CDC42EP2' 0.036 'KIAA1466' 0.011 'SUZ12' 0.026 'A_24_P24786' 0.048 'CHMP4B' 0.016 'KIAA1505' 0.020 'TCF20' 0.019 'A_24_P289834' 0.033 'CIDEC' 0.044 'KLHL23' 0.028 'THC2305027' 0.023 'A_24_P298099' 0.042 'CLEC4A' 0.007 'KRT9' 0.011 'THC2309459' 0.038 'A_24_P340866' 0.024 'COQ10A' 0.038 'LOC133874' 0.040 'THC2315164' 0.009 'A_24_P375405' 0.011 'CR591289' 0.011 'LOC222171' 0.007 'THC2337923' 0.041 'A_24_P384239' 0.016 'CR598370' 0.031 'LOC256021' 0.022 'THC2339658' 0.043 'A_24_P410256' 0.016 'CR601694' 0.047 'LOC283551' 0.023 'THC2339776' 0.040 'A_24_P67552' 0.030 'CR609948' 0.029 'LOC285033' 0.019 'THC2373805' 0.009 'A_24_P681563' 0.035 'CR610211' 0.025 'LOC286495' 0.033 'THC2377297' 0.017 'A_24_P850187' 0.036 'CR612065' 0.049 'LOC644656' 0.045 'THC2378276' 0.039 'A_24_P88493' 0.049 'CR624880' 0.047 'LTC4S' 0.039 'THC2405400' 0.016 'A_32_P107797' 0.016 'DLST' 0.046 'MAFG' 0.009 'THC2406815' 0.041 'A_32_P127412' 0.045 'DNAJA5' 0.041 'MAGEF1' 0.038 'THC2407189' 0.044 'A_32_P147603' 0.036 'DNASE1L3' 0.035 'MAPBPIP' 0.043 'THC2438117' 0.026 'A_32_P225768' 0.031 'DQX1' 0.026 'MAX' 0.042 'THC2438994' 0.022 'A_32_P9924' 0.035 'DUSP14' 0.008 'MFHAS1' 0.043 'THC2439806' 0.028 'ACY1L2' 0.018 'EDN1' 0.039 'MGC13017' 0.014 'THRAP2' 0.034 'AF086548' 0.028 'ELF2' 0.019 'MGC16121' 0.005 'TIGD7' 0.035 'AF130099' 0.020 'ELMO3' 0.048 'MGC52282' 0.043 'TLR6' 0.021 'AF131777' 0.044 'EMD' 0.008 'MGC70870' 0.011 'TMEM85' 0.026 'AGPAT4' 0.048 'ENST00000270201' 0.019 'MPEG1' 0.049 'TMEM88' 0.024 'AK054718' 0.028 'ENST00000292562' 0.022 'MYO1A' 0.040 'TNFSF14' 0.033 'AK057740' 0.045 'ENST00000297544' 0.039 'MYOM1' 0.029 'TROVE2' 0.036 'AK058000' 0.027 'ENST00000302078' 0.002 'NEK3' 0.026 'UBE2V2' 0.048 'AK091569' 0.050 'ENST00000308621' 0.043 'NR4A3' 0.031 'UBE3A' 0.017 'AK093640' 0.043 'ENST00000319246' 0.044 'NT5E' 0.016 'UBTF' 0.037 'AK095707' 0.022 'ENST00000324982' 0.041 'OR5L2' 0.043 'UGCG' 0.032 'AK096154' 0.030 'ENST00000332804' 0.050 'OXTR' 0.019 'WWP1' 0.021 'AK097322' 0.011 'ENST00000333731' 0.032 'PAPOLG' 0.003 'YTHDC2' 0.042 'AK127494' 0.049 'ENST00000339367' 0.027 'PFDN6' 0.042 'ZCCHC2' 0.028 'ALS2' 0.043 'ENST00000342688' 0.033 'PGLYRP4' 0.047 'ZMAT2' 0.047 'APBB1IP' 0.028 'ENST00000356126' 0.020 'PHACTR4' 0.017 'ZNF121' 0.038 'ARHGAP26' 0.046 'ENST00000361227' 0.007 'PHLDA2' 0.042 'ZNF302' 0.016 'ATP6V1H' 0.042 'ENST00000361567' 0.004 'PHTF1' 0.041 'ZNF589' 0.049 'AV714556' 0.041 'ENST00000381108' 0.049 'PIN4' 0.041 'ZNF642' 0.050 'BAGE' 0.025 'ENST00000382004' 0.037 'POLR2C' 0.042 'BC010544' 0.041 'EPM2A' 0.006 'POLR2K' 0.036 'BC036361' 0.027 'FAM120AOS' 0.010 'PRKY' 0.006 'BC040991' 0.044 'FBXO34' 0.012 'PRRG4' 0.017 'BC047952' 0.023 'FCN1' 0.024 'PTPN11' 0.035 'BC054888' 0.009 'FGF2' 0.030 'PYCR2' 0.042 'BC066989' 0.004 'FLJ10099' 0.025 'RABGGTB' 0.040 'BF576096' 0.040 'FLJ21127' 0.048 'RALGDS' 0.042 'BQ050540' 0.033 'FLJ22639' 0.050 'RCC2' 0.020 'BU56609’ 0.032 'FLJ31951' 0.046 'RDH11' 0.042 'BU566406' 0.018 'FLJ37453' 0.023 'RKHD1' 0.048 'BU661610' 0.026 'FLJ39779' 0.021 'RNF121' 0.039 'C14orf128' 0.007 'GABARAPL2' 0.044 'RWDD3' 0.042 'C15orf27' 0.030 'GIMAP8' 0.035 'S100A10' 0.043 'C1orf182' 0.023 'GPR65' 0.029 'SART2' 0.009 'C21orf119' 0.014 'HIRA' 0.037 'SDHD' 0.021 'C3orf17' 0.030 'HNRPH3' 0.037 'SECISBP2' 0.040 'C7orf28A' 0.034 'HOMER1' 0.040 'SHOX2' 0.012 *Features "called" more than once only represented once in table, total # called features = 476 (up & down)

106

TABLE S-T3: BRCA2 IR Classifier: Down-Regulated Features (236)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P151376' 0.033 'CDCA3' 0.049 'LEF1' 0.019 'RP11-308D16.4' 0.046 'A_24_P471121' 0.037 'CDCA7' 0.018 'LHFPL4' 0.024 'RP4-742C19.3' 0.027 'A_24_P565898' 0.007 'CENPI' 0.032 'LOC133619' 0.036 'RRM2' 0.039 'A_24_P584463' 0.042 'CENPL' 0.011 'LOC134147' 0.025 'SAAL1' 0.046 'A_24_P67494' 0.042 'CENPO' 0.049 'LOC162427' 0.029 'SAP30L' 0.044 'A_24_P732439' 0.044 'CENPQ' 0.040 'LOC63929' 0.045 'SCAMP1' 0.024 'A_24_P84711' 0.022 'CERK' 0.034 'LOC643513' 0.030 'SEC14L1' 0.050 'A_24_P931711' 0.016 'CHP' 0.012 'LRRFIP2' 0.028 'SEC22C' 0.021 'A_32_P108748' 0.006 'CIT' 0.029 'LY6G5C' 0.029 'SFXN2' 0.019 'AATK' 0.024 'CKAP4' 0.047 'MANEAL' 0.028 'SGOL1' 0.020 'ACBD3' 0.024 'CR600872' 0.045 'MECP2' 0.021 'SLC16A7' 0.030 'ADAM33' 0.043 'CR625942' 0.050 'METTL2B' 0.028 'SLC25A30' 0.024 'ADD1' 0.046 'CRBN' 0.034 'MGC4473' 0.008 'SLC31A1' 0.011 'AI084055' 0.037 'CRTAP' 0.025 'MIZF' 0.026 'SLC33A1' 0.011 'AK027173' 0.010 'CTCF' 0.013 'MPZL1' 0.041 'SLC39A8' 0.028 'ALG10' 0.025 'DENND1C' 0.024 'MRPL43' 0.048 'SLIT3' 0.003 'ALG10B' 0.048 'DKFZP564O0523' 0.026 'MUTYH' 0.046 'SMARCE1' 0.023 'ALG9' 0.023 'DKFZp762E1312' 0.026 'NDST1' 0.031 'SPATS2' 0.037 'AMFR' 0.036 'DNMT3A' 0.014 'NDUFA9' 0.015 'SPIN' 0.041 'ANKHD1' 0.036 'DYRK4' 0.040 'NEK2' 0.001 'SPRR3' 0.011 'ARHGAP11A' 0.015 'EDEM2' 0.020 'NFYC' 0.034 'SPTLC2' 0.018 'ARL5B' 0.018 'EIF2C2' 0.032 'NOS3' 0.048 'ST3GAL5' 0.020 'ARL6IP' 0.036 'ELF1' 0.040 'NR2C2' 0.003 'STAMBP' 0.028 'ARPC5L' 0.025 'ELF4' 0.024 'NSMCE2' 0.033 'STYX' 0.001 'ASB8' 0.011 'EME1' 0.024 'NUDT2' 0.031 'THC2268216' 0.039 'ATF2' 0.047 'ENST00000217537' 0.042 'NUP205' 0.037 'THC2308298' 0.045 'AW235815' 0.042 'ENST00000279968' 0.041 'NY-SAR-48' 0.007 'THC2364893' 0.046 'AW291149' 0.016 'ENST00000312943' 0.026 'ODF2' 0.018 'THC2408500' 0.012 'B4GALNT1' 0.002 'ENST00000314295' 0.013 'OGT' 0.030 'TICAM2' 0.018 'BAIAP2L1' 0.022 'ENST00000379285' 0.026 'OR7E13P' 0.024 'TIMELESS' 0.003 'BC042649' 0.028 'ENST00000380874' 0.032 'OR7E47P' 0.043 'TMED7' 0.002 'BHLHB8' 0.007 'ENST00000382990' 0.025 'ORC1L' 0.022 'TMEM111' 0.025 'BRCA2' 0.024 'EXO1' 0.014 'ORC6L' 0.034 'TNFRSF10B' 0.002 'BX346853' 0.036 'FAM104A' 0.039 'OSBP' 0.036 'TNFSF5IP1' 0.001 'C11orf17' 0.046 'FAM64A' 0.044 'P2RX3' 0.044 'TNKS' 0.020 'C14orf130' 0.046 'FAM76A' 0.046 'PARN' 0.007 'TRIB1' 0.037 'C14orf133' 0.011 'FAM98A' 0.045 'PEX11B' 0.014 'TRIM22' 0.040 'C14orf140' 0.047 'FANCG' 0.047 'PEX19' 0.044 'TRIM4' 0.004 'C14orf65' 0.003 'FBXO4' 0.043 'PGEA1' 0.047 'TRIM5' 0.031 'C15orf40' 0.004 'FLI1' 0.004 'PGRMC2' 0.029 'TTC1' 0.012 'C15orf42' 0.016 'FLJ30596' 0.011 'PKMYT1' 0.024 'TXNDC12' 0.020 'C16orf63' 0.025 'GAS8' 0.048 'PPP1R3F' 0.029 'UGDH' 0.027 'C18orf24' 0.021 'GBP2' 0.015 'PQBP1' 0.049 'UHRF1' 0.037 'C18orf45' 0.020 'GFI1' 0.011 'PRIM2A' 0.038 'UNC84A' 0.045 'C1orf135' 0.004 'GINS3' 0.039 'PRPH' 0.005 'VCY' 0.043 'C1orf96' 0.040 'GOLPH3L' 0.035 'PSME1' 0.047 'WDR19' 0.004 'C20orf30' 0.008 'GOSR2' 0.031 'QRSL1' 0.018 'WDR71' 0.028 'C20orf45' 0.033 'GPATC1' 0.016 'RAB22A' 0.033 'WRB' 0.005 'C5orf14' 0.034 'GPSM2' 0.001 'RAB6IP2' 0.042 'WWOX' 0.002 'C6orf166' 0.017 'GYG1' 0.047 'RAD51L1' 0.048 'XKR8' 0.016 'CAPZA1' 0.037 'HNRPDL' 0.019 'RB1' 0.027 'ZC3H14' 0.012 'CBFB' 0.047 'HP1BP3' 0.029 'RBM3' 0.047 'ZFP64' 0.004 'CCDC117' 0.043 'INPP5A' 0.025 'RBMX2' 0.009 'ZFYVE26' 0.017 'CCNA2' 0.042 'KDELR2' 0.002 'RFC5' 0.036 'ZMYM6' 0.047 'CCNT2' 0.044 'KIAA0241' 0.048 'RNASE4' 0.009 'ZNF473' 0.041 'CCR5' 0.016 'LARP2' 0.022 'RNF149' 0.025 'ZNF586' 0.043 'CDC42SE2' 0.027 'LAYN' 0.044 'RNF36' 0.038 'ZNF587' 0.028 *Features "called" more than once only represented once in table, total # called features = 476 (up & down)

107

TABLE S-T4: BRCA1 IR-AD Classifier: Up-Regulated Features (406)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P9509' 0.041 'AF070529' 0.030 'C14orf156' 0.040 'EMD' 0.045 'A_24_P101211' 0.046 'AF132973' 0.030 'C14orf2' 0.012 'ENST00000224809' 0.025 'A_24_P101271' 0.037 'AF271776' 0.016 'C15orf27' 0.016 'ENST00000242224' 0.012 'A_24_P110082' 0.027 'AF289590' 0.029 'C16orf61' 0.048 'ENST00000244249' 0.037 'A_24_P110601' 0.038 'AGER' 0.043 'C19orf24' 0.016 'ENST00000289105' 0.045 'A_24_P118382' 0.045 'AGMAT' 0.020 'C19orf39' 0.019 'ENST00000297544' 0.023 'A_24_P127063' 0.029 'AI079353' 0.011 'C19orf46' 0.022 'ENST00000297812' 0.028 'A_24_P161317' 0.036 'AI216457' 0.030 'C1orf162' 0.012 'ENST00000298453' 0.042 'A_24_P204165' 0.049 'AI608782' 0.016 'C1orf54' 0.041 'ENST00000299756' 0.021 'A_24_P212764' 0.043 'AI916628' 0.017 'C1orf57' 0.026 'ENST00000307366' 0.027 'A_24_P229438' 0.048 'AIF1' 0.012 'C20orf4' 0.012 'ENST00000308989' 0.032 'A_24_P230457' 0.021 'AK024035' 0.022 'C20orf6' 0.050 'ENST00000309884' 0.027 'A_24_P238836' 0.037 'AK027319' 0.034 'C21orf119' 0.007 'ENST00000312412' 0.046 'A_24_P24786' 0.021 'AK057740' 0.035 'C22orf16' 0.018 'ENST00000313760' 0.030 'A_24_P25040' 0.015 'AK097322' 0.029 'C3orf15' 0.030 'ENST00000316294' 0.048 'A_24_P298029' 0.043 'AK126405' 0.021 'C3orf54' 0.043 'ENST00000316577' 0.043 'A_24_P32930' 0.029 'AK2' 0.010 'C5orf24' 0.044 'ENST00000329309' 0.037 'A_24_P332263' 0.034 'AP2S1' 0.024 'C5orf4' 0.017 'ENST00000329627' 0.048 'A_24_P375132' 0.033 'APOL6' 0.026 'C6orf148' 0.032 'ENST00000331419' 0.016 'A_24_P383834' 0.048 'ARHGAP25' 0.019 'C6orf149' 0.027 'ENST00000339867' 0.043 'A_24_P391748' 0.022 'ARIH2' 0.027 'C6orf49' 0.047 'ENST00000341824' 0.039 'A_24_P409361' 0.009 'ARL17P1' 0.019 'C6orf52' 0.025 'ENST00000342294' 0.035 'A_24_P41781' 0.019 'ASCC3' 0.022 'C6orf57' 0.030 'ENST00000342688' 0.044 'A_24_P477353' 0.031 'ASRGL1' 0.048 'C9orf123' 0.017 'ENST00000345365' 0.024 'A_24_P600603' 0.013 'ASTN2' 0.000 'C9orf3' 0.025 'ENST00000355629' 0.025 'A_24_P635355' 0.028 'ATOX1' 0.008 'C9orf9' 0.047 'ENST00000357778' 0.041 'A_24_P681563' 0.049 'ATXN7L2' 0.005 'CABYR' 0.014 'ENST00000361227' 0.005 'A_24_P733083' 0.047 'AW075437' 0.006 'CB993898' 0.044 'ENST00000361681' 0.020 'A_24_P7494' 0.036 'AW168145' 0.026 'CCL17' 0.015 'ENST00000366699' 0.043 'A_24_P807445' 0.044 'AW327568' 0.014 'CCL5' 0.015 'ENST00000372866' 0.003 'A_24_P84268' 0.022 'AW804491' 0.040 'CCNK' 0.016 'ENST00000376068' 0.020 'A_24_P843552' 0.040 'AY170823' 0.008 'CDC42' 0.026 'ENST00000379855' 0.039 'A_24_P868905' 0.005 'AY562498' 0.046 'CEACAM1' 0.047 'EP300' 0.039 'A_24_P884915' 0.030 'AYP1p1' 0.042 'CF528315' 0.034 'FAM111B' 0.042 'A_24_P913620' 0.008 'BC000206' 0.020 'CHCHD5' 0.011 'FAM43B' 0.048 'A_32_P10643' 0.026 'BC001764' 0.032 'CK300181' 0.046 'FANCC' 0.024 'A_32_P107797' 0.028 'BC011779' 0.023 'CKLF' 0.040 'FANCF' 0.050 'A_32_P109645' 0.038 'BC011998' 0.032 'CLEC4A' 0.049 'FCER1G' 0.041 'A_32_P112531' 0.019 'BC015588' 0.033 'CMTM4' 0.022 'FIS1' 0.036 'A_32_P147603' 0.010 'BC019044' 0.014 'COL4A5' 0.036 'FKBP1A' 0.042 'A_32_P151782' 0.036 'BC046172' 0.023 'COL5A2' 0.031 'FLJ10803' 0.012 'A_32_P196411' 0.030 'BC054888' 0.013 'COX7B' 0.047 'FLJ10815' 0.033 'A_32_P210744' 0.034 'BC062294' 0.034 'CR597807' 0.048 'FLJ13305' 0.035 'A_32_P35031' 0.047 'BE695822' 0.044 'CR602285' 0.038 'FLJ14346' 0.035 'A_32_P51714' 0.007 'BE769489' 0.041 'CR611723' 0.039 'FLJ14640' 0.040 'A_32_P57002' 0.044 'BF718543' 0.014 'CR626222' 0.029 'FLJ21127' 0.034 'A_32_P64025' 0.043 'BG058000' 0.035 'CRELD1' 0.034 'FOLR2' 0.036 'A_32_P83795' 0.016 'BG775850' 0.020 'CROP' 0.025 'FRMD4A' 0.031 'A_32_P8971' 0.002 'BM677531' 0.039 'CUL5' 0.048 'GCNT1' 0.031 'A_32_P91156' 0.049 'BOLA1' 0.035 'CXorf15' 0.041 'GLIPR1' 0.035 'AA594808' 0.016 'BOLA3' 0.030 'DB379047' 0.050 'GOLGA4' 0.049 'AA725389' 0.018 'BQ000122' 0.040 'DCXR' 0.048 'GPC2' 0.023 'AA856716' 0.028 'BTN2A1' 0.039 'DDX17' 0.025 'GPR84' 0.022 'AA876351' 0.029 'BX100298' 0.005 'DEFB4' 0.026 'HB-1' 0.016 'AA916168' 0.042 'BX352604' 0.040 'DHRS1' 0.048 'HEAB' 0.040 'AA991488' 0.021 'BY798288' 0.035 'DHRS7B' 0.014 'HELZ' 0.042 'AANAT' 0.039 'C10orf6' 0.049 'DHRSX' 0.014 'HEMK1' 0.011 'ABCC4' 0.024 'C14orf119' 0.037 'DIAPH3' 0.031 'HIST1H2AB' 0.049 'ACTR2' 0.038 'C14orf124' 0.006 'ELF2' 0.003 'HIST1H4B' 0.012 *Features "called" more than once are only represented once in table, total # called features = 907 (up & down)

108

TABLE S-T4: BRCA1 IR-AD Classifier: Up-Regulated Features Cont'd (406) Gene IDs p-value Gene IDs p-value Gene IDs p-value 'HIST1H4D' 0.028 'PDCD7' 0.028 'THC2343880' 0.049 'HIST1H4E' 0.016 'PDE4C' 0.048 'THC2345392' 0.028 'HIST2H2AC' 0.047 'PH-4' 0.004 'THC2355570' 0.045 'HNRPA3' 0.040 'PHF2' 0.044 'THC2361211' 0.004 'HSPC023' 0.017 'PHF5A' 0.047 'THC2365651' 0.025 'HYAL3' 0.015 'PHKG2' 0.021 'THC2376140' 0.042 'ID3' 0.004 'PIGL' 0.010 'THC2406815' 0.026 'IFITM2' 0.048 'PNPLA4' 0.039 'THC2407189' 0.043 'IGF2BP1' 0.038 'POLR2H' 0.046 'THC2408033' 0.041 'IL6R' 0.014 'POLR2I' 0.023 'THC2425829' 0.015 'IMP3' 0.023 'POLR2J' 0.024 'THC2434278' 0.030 'ITSN2' 0.043 'POLR2J2' 0.022 'THC2436485' 0.035 'JMJD1C' 0.047 'POLR3G' 0.038 'THC2436745' 0.019 'KCTD9' 0.030 'POM121' 0.041 'THC2437618' 0.039 'KIAA0494' 0.030 'PPM1K' 0.039 'THC2442550' 0.011 'KIAA0738' 0.026 'PRKAB2' 0.043 'THRAP2' 0.030 'KIF21A' 0.044 'PRO2900' 0.024 'TIGD6' 0.047 'LIMK1' 0.037 'PRR4' 0.019 'TIMM17B' 0.042 'LOC133874' 0.032 'PRRG4' 0.015 'TIMP1' 0.020 'LOC147650' 0.020 'PTD008' 0.030 'TM7SF3' 0.046 'LOC150763' 0.035 'PXMP4' 0.033 'TMEM53' 0.018 'LOC205251' 0.004 'RAB5C' 0.019 'TMEM61' 0.047 'LOC254128' 0.041 'RABIF' 0.041 'TMOD4' 0.013 'LOC283887' 0.043 'RANGNRF' 0.032 'TNFSF14' 0.015 'LOC285958' 0.038 'RGC32' 0.034 'TP53RK' 0.022 'LOC374395' 0.027 'RGS12' 0.002 'TP73L' 0.035 'LOC387787' 0.048 'RPE' 0.035 'TPST1' 0.031 'LOC388886' 0.028 'RPP21' 0.043 'TRAPPC2L' 0.043 'LOC388965' 0.017 'RSN' 0.019 'TRERF1' 0.042 'LOC400590' 0.023 'SAC3D1' 0.009 'TUSC2' 0.018 'LOC440288' 0.007 'SCAMP5' 0.039 'U12206' 0.039 'LOC440354' 0.041 'SEPW1' 0.031 'UBE2Q1' 0.008 'LOC51252' 0.043 'SERPINF1' 0.017 'UGT1A6' 0.009 'LSM12' 0.023 'SFRS8' 0.039 'UQCR' 0.023 'Magmas' 0.013 'SH3BP5L' 0.040 'UROD' 0.011 'MAP3K10' 0.011 'SIGIRR' 0.026 'USP51' 0.014 'MAPK11' 0.040 'SLC25A13' 0.039 'VPREB3' 0.009 'MGC16121' 0.034 'SLC7A7' 0.046 'WASF2' 0.029 'MGC21830' 0.044 'SMAD3' 0.017 'WDR59' 0.017 'MGC33556' 0.020 'SMCR7' 0.040 'WDR61' 0.021 'MGC4093' 0.031 'SMG6' 0.023 'WDR90' 0.014 'MOCOS' 0.033 'SPFH2' 0.018 'WFDC2' 0.032 'MRPL2' 0.030 'SPR' 0.042 'WFDC3' 0.031 'MRPL33' 0.037 'ST7L' 0.035 'X62691' 0.024 'MRPL55' 0.046 'SULT1A2' 0.008 'ZAP70' 0.028 'MRPS16' 0.042 'TAF5L' 0.023 'ZC3H6' 0.022 'MST1R' 0.045 'TANC1' 0.049 'ZCCHC2' 0.035 'MYST4' 0.026 'TBP' 0.026 'ZFYVE19' 0.016 'N75427' 0.008 'TBXAS1' 0.021 'ZFYVE9' 0.049 'N91552' 0.033 'tcag7.441' 0.044 'ZNF18' 0.046 'NCR3' 0.020 'TDG' 0.028 'ZNF576' 0.015 'NENF' 0.001 'TGM1' 0.029 'ZNF584' 0.044 'NEXN' 0.034 'THC2265980' 0.041 'NKG7' 0.022 'THC2268341' 0.023 'NOTUM' 0.005 'THC2268405' 0.020 'NUDT22' 0.026 'THC2274051' 0.013 'P117' 0.009 'THC2282095' 0.025 'PANK2' 0.005 'THC2311248' 0.043 'PDCD2' 0.045 'THC2337454' 0.049

109

TABLE S-T4: BRCA1 IR-AD Classifier: Down-Regulated Features (402)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P151376' 0.006 'BCL2L11' 0.048 'CTDSPL2' 0.025 'FLJ11021' 0.040 'A_23_P154006' 0.036 'BE044472' 0.050 'CTNNB1' 0.021 'FLJ21908' 0.010 'A_23_P2032' 0.012 'BM455859' 0.041 'CUTC' 0.026 'FLJ30596' 0.021 'A_24_P144314' 0.039 'BRCA1' 0.036 'CXorf56' 0.045 'FLJ38663' 0.022 'A_24_P230416' 0.028 'BRD1' 0.026 'CXorf57' 0.034 'FNDC3A' 0.032 'A_24_P332532' 0.041 'BRD7' 0.020 'CYP51A1' 0.044 'FOXJ3' 0.013 'A_24_P401670' 0.043 'BRDG1' 0.029 'DC-UbP' 0.036 'FXR1' 0.033 'A_24_P464798' 0.040 'BTBD7' 0.016 'DCUN1D4' 0.047 'GALNT1' 0.032 'A_24_P474188' 0.014 'BXDC2' 0.011 'DDX27' 0.041 'GCA' 0.031 'A_24_P50509' 0.036 'C10orf86' 0.023 'DENND4A' 0.039 'GDI2' 0.020 'A_24_P713893' 0.044 'C11orf46' 0.013 'DENR' 0.023 'GMCL1' 0.018 'A_24_P732439' 0.044 'C14orf111' 0.022 'DHRS7' 0.002 'GMPS' 0.007 'A_24_P75688' 0.039 'C14orf138' 0.025 'DKFZP564C152' 0.006 'GNB1' 0.035 'A_24_P75718' 0.034 'C15orf15' 0.042 'DKFZP564O0523' 0.045 'GOLGA7' 0.006 'A_24_P780609' 0.030 'C15orf40' 0.046 'DKFZP779L1068' 0.031 'GON4L' 0.049 'A_24_P846755' 0.033 'C17orf27' 0.033 'DRCTNNB1A' 0.038 'GPR109A' 0.031 'A_24_P84873' 0.026 'C17orf80' 0.035 'DUSP11' 0.020 'GPR89A' 0.034 'A_24_P938006' 0.010 'C18orf54' 0.028 'DYNC2LI1' 0.040 'GTF2I' 0.034 'AATF' 0.050 'C1orf119' 0.033 'EDD1' 0.018 'HBLD2' 0.042 'ABHD2' 0.021 'C1orf135' 0.025 'EIF2S3' 0.040 'HECTD1' 0.041 'ABR' 0.039 'C1orf149' 0.009 'EIF4E' 0.048 'HIF1A' 0.036 'ACBD3' 0.013 'C1orf80' 0.017 'ELF4' 0.043 'HIPK1' 0.026 'AEBP2' 0.018 'C20orf30' 0.049 'ELMO2' 0.016 'HNRPDL' 0.050 'AF088033' 0.026 'C20orf45' 0.025 'ENST00000246024' 0.003 'HOOK3' 0.014 'AF445027' 0.025 'C22orf5' 0.009 'ENST00000259969' 0.043 'ICF45' 0.036 'AHCTF1' 0.031 'C3orf37' 0.012 'ENST00000265450' 0.003 'IDI1' 0.031 'AK021546' 0.014 'C5orf15' 0.034 'ENST00000282169' 0.040 'IFIH1' 0.047 'AK026418' 0.029 'C6orf62' 0.022 'ENST00000282964' 0.012 'IFIT2' 0.021 'AK055679' 0.028 'C6orf96' 0.024 'ENST00000299563' 0.032 'IFIT3' 0.027 'AK057591' 0.042 'C9orf5' 0.012 'ENST00000303246' 0.016 'IFNAR1' 0.008 'AK057981' 0.039 'CAPZA1' 0.018 'ENST00000310492' 0.046 'IGF2R' 0.030 'AK074696' 0.014 'CAV1' 0.019 'ENST00000323501' 0.016 'INADL' 0.016 'AK074960' 0.004 'CCDC55' 0.010 'ENST00000325863' 0.049 'INPP5A' 0.011 'AK092090' 0.014 'CCDC98' 0.013 'ENST00000328046' 0.048 'ITPR1' 0.008 'AK092888' 0.014 'CCNG1' 0.044 'ENST00000334464' 0.019 'KCTD21' 0.046 'AK123627' 0.044 'CD2AP' 0.049 'ENST00000354586' 0.047 'KDELR2' 0.039 'AK123765' 0.039 'CDC37L1' 0.036 'ENST00000355854' 0.033 'KIAA0430' 0.012 'AMD1' 0.039 'CDKN1B' 0.017 'ENST00000356730' 0.037 'KIAA1411' 0.046 'ANKRD28' 0.042 'CDV3' 0.034 'ENST00000367740' 0.022 'KIAA1432' 0.005 'ANKRD6' 0.016 'CGGBP1' 0.010 'ENST00000369453' 0.030 'KIAA1450' 0.041 'AP1S3' 0.020 'CGI-09' 0.015 'ENST00000370238' 0.028 'KIAA1715' 0.032 'APC' 0.015 'CHMP2B' 0.011 'ENST00000373219' 0.037 'KIF2' 0.013 'ARHGAP15' 0.019 'CLASP2' 0.046 'ENST00000373954' 0.004 'KIFAP3' 0.025 'ARHGEF12' 0.036 'CLCC1' 0.046 'ENST00000376573' 0.009 'KRAS' 0.017 'ARHGEF18' 0.024 'CLIC4' 0.011 'ENST00000379156' 0.020 'LEF1' 0.025 'ARID1A' 0.005 'CLINT1' 0.022 'ENST00000382108' 0.006 'LNX2' 0.007 'ARIH1' 0.019 'CLPX' 0.027 'ENST00000382990' 0.035 'LOC129285' 0.038 'ATP2C1' 0.020 'CMPK' 0.045 'EPN1' 0.039 'LOC153346' 0.011 'ATP6V1A' 0.023 'CNOT2' 0.019 'ESCO1' 0.037 'LOC285535' 0.019 'ATP6V1C1' 0.020 'CNOT6' 0.015 'ESR2' 0.015 'LOC389831' 0.032 'ATP6V1D' 0.008 'COG3' 0.029 'EXOD1' 0.034 'LOC51136' 0.023 'ATRX' 0.020 'COPA' 0.048 'EZH2' 0.038 'LOC57149' 0.049 'AW663344' 0.035 'CPNE8' 0.019 'FAM117A' 0.040 'LOC648674' 0.035 'BARD1' 0.050 'CREBBP' 0.009 'FAM18B2' 0.030 'LOC90624' 0.030 'BBS7' 0.034 'CRLF3' 0.047 'FAM3C' 0.010 'LRRC42' 0.044 'BC014384' 0.040 'CRSP6' 0.049 'FAM76A' 0.035 'LRRIQ2' 0.023 'BC027347' 0.035 'CSDE1' 0.036 'FAM98A' 0.026 'LSM14A' 0.016 'BC038355' 0.026 'CSNK2A1' 0.036 'FBXW8' 0.017 'MALT1' 0.037 'BC047111' 0.040 'CTCF' 0.033 'FEZ2' 0.049 'MAN2A1' 0.047 *Features "called" more than once are only represented once in table, total # called features = 907 (up & down)

110

TABLE S-T4: BRCA1 IR-AD Classifier: Down-Regulated Features Cont'd (402) Gene IDs p-value Gene IDs p-value Gene IDs p-value 'Mar-07' 0.032521128 'RMND5A' 0.030013709 'THOC1' 0.044138076 'MAT2B' 0.020728097 'RNF149' 0.005342803 'TICAM2' 0.018671876 'MGAT2' 0.048611732 'RNF36' 0.028742278 'TLK1' 0.017660043 'MIER1' 0.01370266 'RNGTT' 0.007666189 'TMED7' 0.009926255 'MKI67IP' 0.046508262 'ROD1' 0.030663168 'TMEM23' 0.021778096 'MOAP1' 0.04713257 'RP11-50D16.3' 0.047004128 'TNFSF5IP1' 0.041124515 'MOBK1B' 0.023173116 'RP2' 0.01746513 'TRAM1' 0.019815752 'MORC3' 0.013046266 'RPL7L1' 0.0329212 'TRIM22' 0.040117384 'MORF4' 0.02275974 'RRN3' 0.033105517 'TRIM4' 0.010951064 'MORF4L1' 0.007835492 'RSL1D1' 0.013746576 'TRIM5' 0.046977246 'MRPL42' 0.019838557 'RUSC2' 0.037078555 'TRIM59' 0.029010294 'MRPL44' 0.022381365 'SAAL1' 0.047141749 'TROVE2' 0.000229114 'MTPN' 0.031250037 'SAT' 0.049961862 'TSPAN3' 0.040456648 'NAT12' 0.036989517 'SBNO1' 0.019311431 'TTC1' 0.039187656 'NIPA1' 0.045243766 'SCML1' 0.033209866 'TUBE1' 0.025908968 'NMT2' 0.021015928 'Sep-02' 0.022176245 'TXNDC4' 0.039781655 'NOC3L' 0.03646829 'SERP1' 0.032724541 'UBE2B' 0.030351877 'OCIAD1' 0.049228669 'SERPINB9' 0.019885642 'UBE2D3' 0.02319603 'P18SRP' 0.021098522 'SHPRH' 0.01615339 'UBE2H' 0.046207179 'PAG1' 0.030924995 'SIN3A' 0.047864365 'UBE3A' 0.01964727 'PAIP2' 0.046570813 'SLA/LP' 0.031995231 'UFM1' 0.015648287 'PAK2' 0.025377934 'SLAMF6' 0.009731164 'UNC84A' 0.015647675 'PARN' 0.008995102 'SLC2A5' 0.019441869 'USP45' 0.049751855 'PBEF1' 0.030547955 'SLC35A3' 0.007742287 'USP9X' 0.047028051 'PCGF5' 0.048395231 'SLC35B4' 0.030746455 'VMD2L3' 0.038682253 'PCLKC' 0.02494608 'SLC35D2' 0.044038702 'VPS13C' 0.022790788 'PCNP' 0.030745122 'SLC35F2' 0.024121219 'VPS24' 0.013896914 'PCYOX1' 0.030405477 'SLC4A7' 0.02635443 'VPS26B' 0.029702851 'PDCD4' 0.005167746 'SMPD4' 0.046747423 'WBSCR19' 0.023130253 'PGRMC2' 0.043243819 'SNRPB2' 0.032641341 'WDR32' 0.036791269 'PICALM' 0.022913732 'SNTB1' 0.020160879 'WDR55' 0.046226491 'PITPNB' 0.025745216 'SNX3' 0.047812886 'XPA' 0.034388863 'PLDN' 0.041775572 'SOX4' 0.016828194 'XRN1' 0.041301726 'PPID' 0.022617556 'SPAST' 0.018676685 'YEATS4' 0.039689712 'PPM1B' 0.017523828 'SRD5A1' 0.03158244 'ZBTB11' 0.045687741 'PPP1CB' 0.044736757 'SRPK2' 0.030922222 'ZCCHC7' 0.043545133 'PPP1CC' 0.025425566 'ST13' 0.028423833 'ZDHHC23' 0.045183116 'PPP2CA' 0.023920671 'STAG2' 0.0354528 'ZFAND1' 0.004652388 'PPP3CA' 0.019932154 'SWAP70' 0.026459604 'ZFP64' 0.032173556 'PRDX3' 0.04803178 'SYNCRIP' 0.043559996 'ZMYM1' 0.026535332 'PRIM2A' 0.018792433 'TAF11' 0.04015549 'ZMYM6' 0.02150211 'PTER' 0.036470082 'TAF1L' 0.012519214 'ZNF302' 0.019742967 'PUM1' 0.005020302 'TAF6' 0.019711907 'ZNF468' 0.041424139 'QKI' 0.018896304 'TANK' 0.028754754 'ZNF529' 0.029728202 'RAB21' 0.031530419 'TBC1D24' 0.026747953 'ZNF566' 0.046016129 'RAB40B' 0.009468922 'TBL1XR1' 0.027860137 'ZNF572' 0.036415952 'RAB7L1' 0.029661933 'TBRG1' 0.020748036 'ZNF586' 0.035835546 'RABGEF1' 0.021329633 'TCEA1' 0.015236843 'ZNF616' 0.043354396 'RAC1' 0.004627132 'TERF2IP' 0.020369305 'RAD23B' 0.036731578 'THC2250386' 0.010950456 'RAD50' 0.035321896 'THC2274885' 0.041022736 'RANBP9' 0.014565114 'THC2278663' 0.017820962 'RASSF2' 0.027432419 'THC2279548' 0.007139093 'RB1' 0.007099938 'THC2281165' 0.030738774 'RBM3' 0.040551046 'THC2302075' 0.0091831 'RFFL' 0.003835152 'THC2313495' 0.012478869 'RHEB' 0.006362959 'THC2397609' 0.008444669 'RHOQ' 0.032197038 'THC2397697' 0.046482221 'RIF1' 0.049834121 'THC2400533' 0.01019841

111

TABLE S-T5: BRCA2 IR-AD Classifier: Up-Regulated Features (221)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_24_P178167' 0.007 'CCBL1' 0.030 'HLA-G' 0.009 'RNF121' 0.011 'A_24_P229903' 0.012 'CCDC26' 0.014 'HRASLS3' 0.019 'SART2' 0.015 'A_24_P289834' 0.017 'CEP290' 0.037 'HTR7P' 0.020 'SCGB3A1' 0.015 'A_24_P307046' 0.030 'CHMP4B' 0.006 'ID3' 0.017 'SCLY' 0.038 'A_24_P340866' 0.038 'CHST2' 0.038 'INA' 0.030 'SDC1' 0.030 'A_24_P375405' 0.046 'CIDEC' 0.039 'ITIH1' 0.040 'SECISBP2' 0.017 'A_24_P375870' 0.049 'CLEC4A' 0.008 'KAZALD1' 0.040 'SHOX2' 0.030 'A_24_P384239' 0.036 'COG8' 0.039 'KIAA0082' 0.031 'SLC4A11' 0.032 'A_24_P410256' 0.014 'COL1A1' 0.026 'KIAA0376' 0.022 'SLC5A6' 0.046 'A_24_P936051' 0.044 'COQ10A' 0.017 'KIAA1466' 0.015 'SLC6A4' 0.012 'A_32_P127412' 0.029 'CR591289' 0.040 'KIAA1505' 0.005 'SORBS3' 0.012 'A_32_P147603' 0.049 'CR598370' 0.036 'KIAA1787' 0.025 'SOX4' 0.022 'A_32_P225768' 0.026 'CR601496' 0.027 'KIAA1908' 0.049 'SYK' 0.018 'A_32_P53670' 0.021 'CR609948' 0.023 'KIAA2010' 0.028 'TCF20' 0.034 'A_32_P61132' 0.022 'CR612065' 0.031 'KLHL23' 0.031 'THC2315164' 0.026 'A_32_P64025' 0.049 'CR624880' 0.028 'KPNA6' 0.010 'THC2337923' 0.008 'A_32_P8371' 0.045 'CRYGS' 0.020 'KRT9' 0.006 'THC2339776' 0.024 'ACHE' 0.048 'CSPG2' 0.025 'LOC133874' 0.035 'THC2342255' 0.004 'ACY1L2' 0.026 'CXCL5' 0.042 'LOC222171' 0.012 'THC2377297' 0.011 'AF086548' 0.015 'DHRS1' 0.024 'LOC285033' 0.006 'THC2383106' 0.028 'AF131777' 0.040 'DLST' 0.013 'LOC440350' 0.031 'THC2384784' 0.030 'AHCTF1' 0.008 'DNAJC7' 0.025 'MAFG' 0.046 'THC2405400' 0.031 'AK058000' 0.028 'DQX1' 0.010 'MAP1LC3A' 0.034 'THC2406815' 0.040 'AK095727' 0.028 'DUSP14' 0.003 'MAP3K10' 0.047 'THC2438117' 0.032 'AK096154' 0.041 'DUSP22' 0.033 'MAPBPIP' 0.049 'THC2439806' 0.042 'AK097322' 0.014 'EMD' 0.002 'MARVELD1' 0.027 'THRAP2' 0.041 'ALDH4A1' 0.020 'ENST00000219169' 0.028 'MAX' 0.028 'TIGD7' 0.020 'ALS2' 0.027 'ENST00000227451' 0.029 'MFHAS1' 0.020 'TMEM54' 0.048 'AP1S1' 0.047 'ENST00000270201' 0.004 'MGC13017' 0.032 'TROVE2' 0.046 'APTX' 0.049 'ENST00000292562' 0.008 'MGC16121' 0.009 'TRPM5' 0.048 'ARHGEF9' 0.039 'ENST00000297544' 0.044 'MGC34796' 0.041 'TTC18' 0.030 'ASTN2' 0.001 'ENST00000302078' 0.004 'MGC52282' 0.002 'TTC21B' 0.045 'ATP6AP1' 0.018 'ENST00000324982' 0.032 'MGC70870' 0.002 'TTC28' 0.041 'ATP6V1H' 0.026 'ENST00000327423' 0.047 'MPEG1' 0.023 'TUSC2' 0.040 'AY203928' 0.048 'ENST00000328419' 0.015 'MRM1' 0.012 'UBE2Q1' 0.042 'AY358802' 0.022 'ENST00000344771' 0.042 'MYO1A' 0.036 'UBTF' 0.014 'BAGE' 0.035 'ENST00000361227' 0.007 'MYOM1' 0.017 'VASH1' 0.012 'BC019824' 0.044 'ENST00000361567' 0.007 'NICN1' 0.031 'VPS45A' 0.037 'BC040991' 0.037 'ENST00000369911' 0.049 'NPR2' 0.044 'WDR63' 0.012 'BC054888' 0.016 'ENST00000370660' 0.045 'NT5E' 0.002 'WWP1' 0.013 'BC066989' 0.043 'ENST00000377548' 0.024 'NXF3' 0.038 'YTHDC2' 0.043 'BCAS3' 0.035 'EPM2A' 0.039 'PAPOLG' 0.033 'ZAP70' 0.027 'BF576096' 0.045 'FAM20B' 0.033 'PFDN6' 0.039 'ZDHHC11' 0.038 'BQ050540' 0.046 'FBXO34' 0.008 'PGLYRP4' 0.026 'ZFYVE9' 0.049 'BU566406' 0.020 'FBXO44' 0.022 'PHACTR4' 0.019 'ZMAT2' 0.034 'BX100298' 0.041 'FCGBP' 0.024 'PHKG2' 0.044 'ZNF589' 0.023 'BX409884' 0.021 'FGF2' 0.020 'PHTF1' 0.022 'C11orf68' 0.050 'FLJ21127' 0.026 'PLCL2' 0.048 'C12orf11' 0.044 'FLJ30092' 0.047 'POLR2K' 0.039 'C14orf128' 0.003 'FLJ37453' 0.024 'POLRMT' 0.042 'C15orf27' 0.006 'FLJ40176' 0.022 'PRKY' 0.014 'C16orf5' 0.032 'GEFT' 0.039 'PTD008' 0.041 'C19orf39' 0.038 'GIMAP8' 0.041 'PYCR2' 0.036 'C19orf6' 0.042 'GPR25' 0.050 'PYGO2' 0.035 'C1orf182' 0.027 'HARS' 0.027 'RAB24' 0.031 'C5orf16' 0.045 'HEMK1' 0.045 'RAD50' 0.042 'C7orf30' 0.037 'HES2' 0.030 'RGPD2' 0.038 'C8orf38' 0.032 'HIRA' 0.039 'RKHD1' 0.042 *Features "called" more than once are only represented once in table, total # called features = 502 (up & down)

112

TABLE S-T5: BRCA2 IR-AD Classifier: Down-Regulated Features (240)* Gene IDs p-value Gene IDs p-value Gene IDs p-value Gene IDs p-value 'A_23_P151376' 0.035 'C5orf25' 0.038 'GBP2' 0.015 'RAB40B' 0.029 'A_24_P110591' 0.037 'CAMKK2' 0.037 'GFI1' 0.046 'RAB5A' 0.049 'A_24_P205307' 0.038 'CAPZA1' 0.010 'GINS3' 0.031 'RAB6IP2' 0.028 'A_24_P279760' 0.046 'CAST' 0.045 'GOLPH3L' 0.011 'RAD51L1' 0.018 'A_24_P289565' 0.038 'CCDC117' 0.016 'GOSR2' 0.009 'RBM3' 0.048 'A_24_P307075' 0.040 'CDC2' 0.023 'GPATC1' 0.039 'RBMX2' 0.019 'A_24_P332911' 0.032 'CDC42SE2' 0.045 'GPSM2' 0.000 'RFC5' 0.029 'A_24_P341106' 0.015 'CDC6' 0.036 'HMGB1' 0.040 'RIOK3' 0.023 'A_24_P375573' 0.049 'CDCA7' 0.037 'HNRPDL' 0.006 'RKHD2' 0.036 'A_24_P409521' 0.050 'CDRT4' 0.050 'HP1BP3' 0.014 'RNASE4' 0.023 'A_24_P455100' 0.036 'CENPI' 0.037 'HSA9761' 0.022 'RNF149' 0.005 'A_24_P499215' 0.041 'CENPL' 0.024 'INPP5A' 0.036 'RP11-11C5.2' 0.010 'A_24_P565898' 0.006 'CENPN' 0.037 'KCNK1' 0.040 'RP11-217H1.1' 0.015 'A_24_P761490' 0.039 'CENPQ' 0.036 'KDELR2' 0.002 'RSL1D1' 0.048 'A_24_P84711' 0.023 'CEP55' 0.036 'KIAA1450' 0.044 'SCAMP1' 0.037 'A_24_P931711' 0.046 'CERK' 0.036 'KNTC2' 0.035 'SEC22C' 0.042 'AA608624' 0.031 'CHP' 0.043 'LARP2' 0.005 'SERP1' 0.035 'AATK' 0.040 'CHRAC1' 0.040 'LEF1' 0.011 'SGOL1' 0.025 'ACBD3' 0.005 'CIT' 0.009 'LENG1' 0.035 'SLC25A30' 0.015 'ADAM33' 0.026 'CKAP2' 0.031 'LHFPL4' 0.003 'SLC31A1' 0.001 'ADPRH' 0.036 'CKAP4' 0.045 'LOC133619' 0.010 'SLC33A1' 0.016 'AI084055' 0.048 'CLASP2' 0.048 'LOC399491' 0.038 'SLC39A11' 0.025 'AK027173' 0.008 'CR600872' 0.036 'LOC90321' 0.033 'SLC39A8' 0.005 'ALG10' 0.020 'CRBN' 0.041 'LRRC8A' 0.019 'SLIT3' 0.013 'ALG10B' 0.038 'CREB3L2' 0.050 'LY6G5C' 0.015 'SMARCE1' 0.004 'ALG9' 0.030 'CRTAP' 0.010 'MASTL' 0.049 'SNAP29' 0.041 'ANKHD1' 0.008 'CTCF' 0.030 'MECP2' 0.045 'SPRR3' 0.038 'ANXA7' 0.018 'CUL4A' 0.043 'MED6' 0.026 'SPTLC2' 0.026 'ARHGAP11A' 0.025 'CXorf38' 0.010 'METTL2B' 0.035 'SSR1' 0.040 'ARL5B' 0.004 'CXorf56' 0.026 'MFSD2' 0.040 'STCH' 0.037 'ARL6IP' 0.020 'DDX42' 0.024 'MSN' 0.049 'STYX' 0.000 'ARPC5L' 0.016 'DEGS1' 0.046 'MTBP' 0.030 'TBPL1' 0.034 'ARSG' 0.018 'DKFZP586P0123' 0.026 'MUTED' 0.049 'TFB1M' 0.028 'ASB8' 0.044 'DNAJB9' 0.036 'NDST1' 0.023 'THC2313495' 0.043 'ATPBD1C' 0.030 'DNMT3A' 0.008 'NDUFA9' 0.027 'THC2364893' 0.033 'AW291149' 0.034 'DYNLL2' 0.020 'NEK2' 0.001 'THC2408500' 0.025 'B4GALNT1' 0.004 'EIF2C2' 0.024 'NR2C2' 0.021 'TIMELESS' 0.026 'BC042649' 0.042 'EIF2C4' 0.035 'NUDT21' 0.029 'TMCC3' 0.044 'BHLHB8' 0.037 'EIF4E' 0.033 'NUPL1' 0.043 'TMED7' 0.001 'BTBD7' 0.013 'EIF4EBP2' 0.021 'NUSAP1' 0.035 'TMEM111' 0.043 'BTF3' 0.030 'ELF1' 0.028 'NY-SAR-48' 0.010 'TMEM16H' 0.045 'BUD31' 0.031 'ELF4' 0.006 'ODF2' 0.042 'TNFSF5IP1' 0.003 'C10orf88' 0.047 'ENST00000217537' 0.014 'P2RX3' 0.030 'TNKS' 0.021 'C14orf130' 0.026 'ENST00000236199' 0.045 'PACSIN1' 0.039 'TRIM4' 0.011 'C14orf145' 0.021 'ENST00000311040' 0.048 'PAFAH1B2' 0.022 'TTC1' 0.004 'C14orf65' 0.002 'ENST00000314295' 0.006 'PARN' 0.009 'TXNDC12' 0.027 'C15orf40' 0.003 'ENST00000373954' 0.026 'PDCD6IP' 0.019 'UHRF1' 0.048 'C15orf42' 0.020 'ENST00000376573' 0.042 'PEX11B' 0.024 'UNC84A' 0.039 'C16orf63' 0.009 'ENST00000378954' 0.037 'PEX13' 0.044 'USP6NL' 0.047 'C18orf24' 0.039 'ENST00000382990' 0.013 'PEX19' 0.022 'VCY' 0.034 'C18orf45' 0.022 'EXO1' 0.007 'PGRMC2' 0.011 'WDR19' 0.023 'C1orf135' 0.003 'FAM104A' 0.024 'PIAS2' 0.004 'WDR70' 0.030 'C1orf96' 0.012 'FAM53C' 0.029 'PIGA' 0.030 'WDR71' 0.046 'C20orf12' 0.044 'FAM64A' 0.022 'PPFIBP1' 0.030 'WIPI2' 0.050 'C20orf30' 0.030 'FBXO4' 0.037 'PPP1R3F' 0.032 'WRB' 0.024 'C20orf45' 0.045 'FLI1' 0.015 'PPP3CC' 0.046 'WWOX' 0.016 'C22orf13' 0.044 'FLJ12986' 0.041 'PRKRIP1' 0.026 'ZC3H14' 0.008 'C5orf14' 0.038 'FLJ30596' 0.008 'PRPH' 0.005 'ZFP64' 0.009 'C5orf15' 0.024 'GATAD1' 0.036 'PRPS2' 0.026 'ZMYM6' 0.039 *Features "called" more than once are only represented once in table, total # called features = 502 (up & down)

113

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) ASTN2 12.07 ENST00000340381 6.45 AA887631 5.38 IL27 4.82 SULF2 10.92 TNFSF4 6.40 TNFRSF1A 5.38 CB250445 4.81 THC2429167 10.41 MGC70863 6.40 GLT8D3 5.38 GPR172B 4.80 CABYR 9.64 CN430223 6.38 CDK5R2 5.37 TAAR5 4.80 TM7SF3 9.44 ZNF385 6.35 MAMDC4 5.36 PDGFRA 4.79 HES2 9.42 ASCC3 6.34 GLTSCR1 5.35 TMEM35 4.79 PLK2 9.24 DLGAP4 6.33 LRDD 5.34 ISG20L1 4.79 ANKRD47 9.17 CNGB1 6.32 A_32_P12282 5.33 SESN2 4.78 THC2343350 9.17 BC053363 6.28 DCUN1D3 5.30 A_32_P108748 4.78 A_32_P52153 9.16 DIRC1 6.24 STK4 5.30 CSNK1D 4.76 MDM2 9.09 THC2273762 6.20 PPM1D 5.29 PSTPIP2 4.75 VWCE 9.07 THC2266610 6.17 ALOX5 5.29 ENST00000294485 4.73 BC040303 8.84 BTBD14A 6.15 PARD6G 5.27 C11orf24 4.73 SESN1 8.72 C1orf42 6.13 A_32_P139021 5.26 CR590071 4.72 LIF 8.58 POLH 6.10 LOC441245 5.21 NDUFV3 4.72 AI500335 8.57 C20orf161 6.09 FLJ13576 5.19 AK123704 4.71 GIPR 8.52 MAP4K4 6.08 MARVELD3 5.19 A_23_P64962 4.71 TRIM22 8.51 TNFRSF10B 6.07 BF718543 5.18 LOC253981 4.70 TNFRSF10C 8.48 WDR63 6.06 LENG8 5.18 PIGR 4.70 SMPD3 8.28 SPARC 6.04 FAM84A 5.17 LOC51252 4.69 TRIAP1 8.19 STARD4 6.01 THC2289056 5.15 THC2376737 4.69 LOC134147 8.05 OR11A1 5.96 AK026194 5.14 AK056245 4.68 AK092083 8.00 DUSP18 5.94 LONPL 5.13 PLEKHQ1 4.67 IL10RB 7.99 THC2314369 5.94 C19orf46 5.13 A_32_P225768 4.67 PAPLN 7.99 C1orf57 5.91 CHST6 5.10 PROCR 4.65 TP53I3 7.98 TP53AP1 5.91 SPR 5.09 BI029121 4.65 C12orf5 7.93 ITPKC 5.91 DCP1B 5.08 C1orf102 4.64 GLS2 7.92 ACTA2 5.91 PLK3 5.07 FCER1G 4.64 MGC5370 7.91 MYO1A 5.90 HEMK1 5.07 PODXL 4.63 C1orf183 7.89 ADRB2 5.88 ENST00000368491 5.06 FLJ10815 4.62 NTN1 7.85 PHLDA3 5.86 SDC1 5.06 BC030100 4.62 AK024898 7.74 FAM98C 5.86 APBA3 5.05 ITGAM 4.62 THC2437069 7.64 GFAP 5.83 GRIN2C 5.04 ELF5 4.61 CEACAM1 7.52 A_23_P6514 5.83 TNFRSF10A 5.03 THC2345956 4.60 SLC6A19 7.51 FKSG2 5.79 FUCA1 5.02 MICB 4.60 DB518505 7.48 THC2340838 5.77 FOSL1 5.01 A_24_P144487 4.60 PRSS36 7.46 LOC133874 5.76 SLC30A3 5.01 SLC4A11 4.59 MDS025 7.46 LOC653374 5.76 BRMS1L 4.98 CCDC92 4.57 SPATA18 7.39 GRHL3 5.73 BF960555 4.98 AHRR 4.57 RHO 7.24 BTBD14B 5.68 REEP2 4.97 ENST00000360523 4.56 SARDH 7.20 FEZ1 5.67 KCTD1 4.96 S75896 4.56 FBXO22 7.18 PGPEP1 5.66 CCDC113 4.95 TMPRSS2 4.56 FHL2 7.17 AF144054 5.65 SLC13A2 4.95 KDELC1 4.55 C8orf38 7.09 ARHGEF3 5.64 DQX1 4.95 LOC339768 4.54 PLAT 7.08 RGS12 5.57 LPHN1 4.95 LOC92017 4.54 THC2315966 7.03 EFNB1 5.56 TP53INP1 4.94 PPFIBP2 4.54 BLOC1S2 6.99 C3orf23 5.56 GADD45A 4.94 BM989484 4.54 CFD 6.90 ANKRA2 5.54 TANC1 4.93 THC2440818 4.53 FLJ11259 6.88 GDF15 5.54 A_23_P206568 4.92 C18orf56 4.52 BBC3 6.84 DOCK4 5.48 THC2238625 4.88 NIN 4.51 THC2316748 6.84 SMTN 5.47 THC2346713 4.88 C20orf12 4.51 LOC653483 6.79 FAM41C 5.46 Ells1 4.87 A_23_P207049 4.51 ZNF79 6.69 THC2280109 5.46 BBS4 4.86 ZNF219 4.51 THC2322041 6.64 RRAD 5.44 THC2408506 4.86 A_24_P127042 4.50 AK026338 6.59 ARHGEF15 5.44 A_24_P635355 4.86 A_24_P531074 4.50 HGS 6.56 JUND 5.43 PRKAB1 4.85 COL5A2 4.49 ENST00000377836 6.56 THC2406514 5.41 KRTAP5-8 4.83 THC2311946 4.49 SLC7A6 6.48 GCH1 5.41 SYTL1 4.82 BC066984 4.48 THC2439499 6.46 FAM83H 5.40 PTP4A1 4.82 DR1 4.48

114

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) MRRF 4.47 SMAD5 4.15 CEP164 3.92 MT1A 3.73 EMX1 4.46 FRK 4.15 A_24_P229728 3.91 C5orf4 3.73 MOSPD1 4.45 TMEM129 4.14 PTPN1 3.91 LOC645431 3.72 C2orf13 4.45 AK093416 4.14 THC2440228 3.91 A_24_P221105 3.72 A_23_P17152 4.43 SFXN5 4.13 THC2277837 3.91 KIAA0513 3.72 OR4D2 4.43 SGOL1 4.13 A_23_P27460 3.90 CASQ1 3.71 C19orf6 4.43 APOBEC3G 4.13 THC2287925 3.89 ENST00000378179 3.71 MICAL-L2 4.42 PCGF3 4.12 ATF3 3.89 SPRR2C 3.71 C6orf173 4.41 MYBPC2 4.12 OR2H1 3.88 ENST00000373335 3.70 FAS 4.41 BX093417 4.11 MSI2 3.88 ENST00000377093 3.70 ENST00000361227 4.40 ITM2A 4.11 THC2450504 3.87 CD511705 3.69 TSHB 4.40 CD242823 4.11 KRT17 3.87 LOC158960 3.69 AF086139 4.39 MECP2 4.11 MRPL27 3.86 LOC340061 3.69 CRYZL1 4.39 CDC25C 4.11 PLEKHF1 3.86 NAGS 3.69 RAB8B 4.38 BE646426 4.10 ZNF469 3.86 STOX2 3.69 SERTAD1 4.37 BX101252 4.10 MAP2K6 3.86 MT3 3.68 SAC3D1 4.37 CR626222 4.10 ENST00000316294 3.86 NSF 3.67 TMEM150 4.37 PCGF5 4.09 APBB3 3.86 XPC 3.67 CR609588 4.37 AK026368 4.09 ZNF212 3.86 NOTCH2 3.67 TRIM32 4.37 C16orf5 4.09 A_32_P30187 3.86 TBXA2R 3.66 A_32_P40375 4.37 PLXNB2 4.08 C1orf144 3.86 ZNF30 3.66 HEAB 4.36 THC2310998 4.08 TFR2 3.85 C12orf49 3.66 ST5 4.35 GAMT 4.07 SGPL1 3.85 GRAMD3 3.66 SLC35D1 4.35 TP53TG3 4.07 AADACL1 3.85 LOC647131 3.66 A_24_P383802 4.34 IGSF9 4.06 HNRPH2 3.85 C17orf85 3.66 ARMCX6 4.34 RAB43 4.05 3.8-1 3.84 SLCO2B1 3.65 RRAGA 4.33 THC2437618 4.04 NAP1L5 3.84 ZNF425 3.65 AI916628 4.32 D4ST1 4.04 PLA2G4D 3.84 ZMAT3 3.65 SNX13 4.32 ULK1 4.04 ZNF337 3.84 A_32_P71183 3.65 PRKAB2 4.32 METTL7A 4.03 GM2A 3.82 AY007156 3.65 CYB5R1 4.31 CCL24 4.03 DHTKD1 3.82 RETSAT 3.64 CAPN10 4.31 GAL3ST4 4.02 NDFIP2 3.82 CR603184 3.64 COL4A1 4.29 A_32_P166152 4.02 TRIM35 3.81 TMEM118 3.64 PHF6 4.28 BE138567 4.02 PRIMA1 3.81 HTR7P 3.63 MAGEL2 4.28 PIGH 4.02 SF3A1 3.80 ENST00000335995 3.63 WFDC2 4.28 EI24 4.02 SERPINF1 3.80 AF086329 3.63 PSKH1 4.28 WDR59 4.01 CYP3A5 3.80 MFSD4 3.63 LOC286254 4.28 RASD1 4.00 THC2444653 3.80 CASK 3.62 ENST00000329309 4.28 TRIP6 4.00 FAM43B 3.80 ENST00000361453 3.61 TCP11L1 4.27 OR7E13P 3.99 C1orf26 3.79 THC2248354 3.61 B3GNT8 4.27 OSRF 3.99 A_32_P171348 3.79 AL547361 3.61 FOLR2 4.26 ADAL 3.99 LCE1F 3.79 PEX11B 3.61 THC2314566 4.25 CCNK 3.99 SLC22A5 3.79 BACE1 3.61 NIPSNAP3A 4.25 A_24_P281853 3.98 WDR19 3.79 A_24_P127462 3.61 MLLT1 4.24 SGTB 3.98 UGT2B10 3.78 FCRLM1 3.60 CDKN1A 4.23 PTMS 3.98 SNX22 3.78 A_24_P41781 3.60 ZFP90 4.23 THC2274334 3.97 PGAP1 3.78 IGBP1 3.60 CAND1 4.23 C14orf28 3.96 CREB3L1 3.77 ZNF167 3.60 FGL2 4.23 LOC158863 3.96 SART2 3.76 NCSTN 3.60 LCE2A 4.22 SCGB1A1 3.95 LPXN 3.75 THC2281731 3.60 PHACTR4 4.21 BX111592 3.95 AI076466 3.75 OAT 3.60 ARHGEF10L 4.19 THC2433060 3.95 POMT1 3.75 DEPDC5 3.59 GPR109B 4.19 CRSP3 3.94 ITPR2 3.74 UBR1 3.59 FASTKD5 4.18 THC2376817 3.94 REXO2 3.74 ZFP41 3.59 PDE4C 4.17 STARD8 3.94 ANK1 3.74 LOC644424 3.59 LOC653857 4.17 ORAOV1 3.94 RBM5 3.74 MORN2 3.59 SRA1 4.17 RRM2B 3.94 LOC147650 3.73 ENST00000290607 3.59 STAT4 4.17 THC2243803 3.93 ENST00000373575 3.73 ALCAM 3.59 KIAA0284 4.16 THC2373429 3.93 SLC26A11 3.73 NPAL3 3.59

115

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) BBS1 3.59 ZCWPW1 3.47 ENST00000323198 3.35 UQCR 3.25 LOC646564 3.59 CENTA2 3.47 SLC7A6OS 3.35 BC035751 3.25 BCAS2 3.58 ID3 3.47 POLR2K 3.34 ADH5 3.25 CRHR1 3.58 ENST00000235345 3.47 KIAA1219 3.34 RICS 3.25 BC000206 3.58 FGF3 3.46 C14orf159 3.34 CRYM 3.24 CFL2 3.58 CABC1 3.45 THC2443137 3.34 BAIAP2L1 3.24 CK300181 3.58 A_24_P882309 3.45 ROD1 3.34 RBL2 3.24 OASL 3.58 ZNF134 3.45 ENST00000308118 3.34 OR7E91P 3.24 BM561346 3.57 ASAH2 3.45 FLJ12688 3.33 PCNXL2 3.24 LAX1 3.57 BC035180 3.45 ENST00000299756 3.33 ARL6IP5 3.24 ABHD9 3.57 MRPL48 3.45 ZNF135 3.33 THC2435128 3.24 LOC150763 3.57 ADAM10 3.44 E2F7 3.33 IFI27 3.24 EFCAB4A 3.56 ZNF658 3.44 ALG8 3.33 ABTB2 3.23 TOB1 3.56 EID-3 3.43 ACOT2 3.32 CA13 3.23 TTC12 3.56 C1orf25 3.43 ZNF746 3.32 ZNF655 3.23 LOC442421 3.56 AU185665 3.43 CR745430 3.32 LYK5 3.23 GPT 3.55 ENST00000262525 3.43 RABL2A 3.32 LOC257396 3.23 KIAA0888 3.55 THC2281539 3.43 A_24_P255836 3.32 HSDL2 3.23 MRPL49 3.55 LAT 3.43 THC2381848 3.31 SCNN1D 3.23 ZDHHC7 3.54 MMP19 3.42 AK056119 3.31 ASB7 3.22 VANGL1 3.54 DKFZP434A0131 3.42 MGC23909 3.31 BLCAP 3.22 WFS1 3.54 SLC2A11 3.41 CIAPIN1 3.31 TSPYL3 3.22 PHF16 3.54 KIAA0329 3.41 OR7E156P 3.31 KRTAP10-10 3.22 FLJ10781 3.54 USP33 3.41 HHAT 3.31 MLYCD 3.21 CYB5D2 3.54 MEG3 3.41 ENST00000312289 3.31 AK097322 3.21 THC2290313 3.53 OR10H2 3.41 CYB561D1 3.31 NISCH 3.21 TRAF1 3.53 MR1 3.41 MYH10 3.30 A_24_P868905 3.21 ZCCHC17 3.53 AK092875 3.40 OR7E47P 3.30 HCFC2 3.21 CCNL2 3.53 TEAD3 3.40 THC2427841 3.30 A_24_P144054 3.21 FNDC6 3.53 TRIM62 3.40 A_24_P187365 3.30 ENST00000382004 3.20 CR627133 3.52 SMAD1 3.40 MTERF 3.29 MRPL43 3.20 A_32_P202621 3.52 MTMR3 3.39 HYAL3 3.29 TncRNA 3.20 C9orf90 3.52 A_32_P168727 3.39 ENST00000336283 3.28 FLJ23569 3.20 THC2312748 3.51 MT 3.39 LLGL2 3.28 CR606637 3.20 GGCX 3.51 MGC10471 3.39 LATS2 3.28 ZP3 3.20 BC053632 3.51 SLFN5 3.39 C1orf164 3.28 THEM4 3.20 P117 3.51 LOC647090 3.38 RXRA 3.28 SIN3B 3.20 HGF 3.51 ENST00000298453 3.38 STX1A 3.28 AK055306 3.19 SUMF1 3.50 RBM41 3.38 C4orf24 3.28 LOC649294 3.19 ZNF435 3.50 THC2390306 3.38 CXXC4 3.28 A_24_P75994 3.19 ZMAT2 3.50 MMP28 3.38 HRBL 3.28 OR8B8 3.19 SERPINB8 3.50 KIAA0323 3.38 ZNF707 3.27 C12orf45 3.19 ENST00000372072 3.50 A_24_P913178 3.38 IL1B 3.27 RFFL 3.19 HSPC171 3.49 AL137705 3.38 ATXN3 3.27 ZC3H5 3.19 C11orf41 3.49 PHF15 3.37 THC2442304 3.27 SPOCK2 3.19 OBFC2A 3.49 POP1 3.37 GHSR 3.27 A_24_P25040 3.19 A_32_P151782 3.49 SLC39A13 3.37 BU561469 3.27 FLJ39370 3.19 A_24_P916853 3.49 RP11-217H1.1 3.36 PHF20L1 3.27 MED19 3.19 A_23_P72330 3.49 CASP9 3.36 GYG1 3.27 CCR3 3.18 KIAA0999 3.49 THC2286151 3.36 ENST00000369572 3.27 GRRP1 3.18 PMS2L1 3.49 FLJ32065 3.36 ITM2B 3.26 LOC201229 3.18 CES2 3.48 RBM16 3.36 ITGA7 3.26 CD274 3.18 THC2336837 3.48 CARD12 3.35 FKBP1B 3.26 HSPC048 3.18 RGL1 3.48 ENST00000369739 3.35 LOC550643 3.26 C1RL 3.18 PPARD 3.48 LOC619208 3.35 THC2394812 3.26 RMND5B 3.17 FLJ36868 3.48 RBED1 3.35 PCNX 3.26 RCCD1 3.17 HDAC6 3.47 CTSL2 3.35 TRSPAP1 3.26 A_24_P298604 3.17 CUTL1 3.47 HMGB3 3.35 C20orf142 3.26 SLC22A18AS 3.17 PLEKHA4 3.47 DDX58 3.35 ATG4A 3.25 TMEM142C 3.17

116

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) TSG101 3.17 FMO4 3.06 NPC2 2.99 A_24_P204165 2.92 ENST00000340158 3.16 FZD6 3.06 DND1 2.98 CYP2W1 2.92 ZNF697 3.16 PRKY 3.06 THC2397010 2.98 ZNF253 2.91 AK098422 3.16 PPFIBP1 3.06 DKFZp762I137 2.98 SOLH 2.91 THC2374442 3.16 DHFRL1 3.06 TRIP4 2.98 CR617033 2.91 THC2434943 3.15 AF289566 3.06 FLJ33790 2.98 FGF2 2.91 C1orf88 3.15 SCD5 3.06 MGST2 2.98 ZDHHC16 2.91 FKBP1A 3.15 A_24_P289984 3.06 ARMCX1 2.98 BCL2L14 2.91 A_24_P24890 3.15 A_24_P135551 3.06 C3orf50 2.98 ABHD13 2.91 A_32_P177097 3.15 C17orf59 3.06 ENST00000382592 2.98 GNG4 2.91 ZCCHC10 3.15 AKAP7 3.06 DISP1 2.97 CDC42 2.91 A_24_P170357 3.14 STX6 3.05 BU567832 2.97 KCNK6 2.91 PFKFB2 3.14 AY203961 3.05 ASPM 2.97 THC2281903 2.90 CEP135 3.14 C9orf89 3.05 CDC42EP4 2.97 LOC646626 2.90 DIRC2 3.13 CCDC98 3.05 AVPI1 2.96 LOC653319 2.90 LTBP2 3.13 FLJ10099 3.05 KIAA1005 2.96 OSBPL3 2.90 EXOC7 3.13 ZNF501 3.05 CEP72 2.96 PDCL 2.90 FLJ10241 3.13 BE005242 3.05 THC2439430 2.96 ENST00000380295 2.90 THC2279364 3.13 FAM46A 3.05 COQ10A 2.96 SCYL3 2.89 MDFI 3.12 GNGT2 3.05 VSIG9 2.96 MAP3K12 2.89 CR597807 3.12 ZNF641 3.04 TMEM117 2.96 WDR42A 2.89 A_32_P5628 3.12 AL365520 3.04 ENST00000374860 2.96 EIF2C1 2.89 LOC644053 3.12 MAB21L2 3.04 OTOF 2.96 THC2335955 2.89 THC2284956 3.12 AQP2 3.04 OR7E24 2.96 MAGEE1 2.89 C14orf49 3.12 MMACHC 3.04 A_32_P233569 2.96 HELB 2.89 C5orf5 3.11 TMEM138 3.03 OPN1LW 2.95 JMJD1C 2.89 A_24_P346859 3.11 ACTR1A 3.03 GSK3B 2.95 ENST00000339867 2.88 AK022038 3.11 MAX 3.03 NT5DC1 2.95 MAP3K7IP3 2.88 C9orf19 3.11 COX7B 3.02 GGTL4 2.95 C1D 2.88 A_23_P130639 3.11 A_24_P827794 3.02 C9orf127 2.95 DIP 2.88 CMTM4 3.11 HCP5 3.02 DECR2 2.95 ZNF195 2.88 SCRN3 3.10 PARP3 3.02 DB380193 2.95 TMEM128 2.88 THC2430670 3.10 THC2341675 3.02 CP110 2.95 RY1 2.88 A_24_P375132 3.10 C20orf4 3.02 ZNF498 2.95 PDLIM4 2.88 MYO5A 3.10 NCOA3 3.02 BEX2 2.95 AB055226 2.88 THC2257370 3.10 HUS1B 3.02 SGK 2.95 LRP6 2.87 DPH5 3.10 TTC23 3.01 C16orf58 2.95 AF070529 2.87 LOC285813 3.10 A_24_P341078 3.01 UBPH 2.94 ARHGAP6 2.87 RPS9 3.10 LRRC25 3.01 GGTLA4 2.94 LIPA 2.87 LMO7 3.09 AF281279 3.01 SERTAD2 2.94 PAX4 2.87 ZC3H7A 3.09 RP4-742C19.3 3.01 HDAC5 2.94 ENST00000278949 2.87 THC2300064 3.09 THC2343771 3.01 PRKCDBP 2.94 CASC4 2.87 BSDC1 3.09 TMEM15 3.01 HRAS 2.94 THC2377845 2.87 A_24_P32930 3.09 X75962 3.01 UBQLN1 2.94 C20orf11 2.87 TNFSF10 3.09 PHF8 3.01 MARVELD1 2.94 NOD9 2.87 FKSG44 3.09 AK094477 3.00 JMJD2D 2.94 GPC2 2.87 MPEG1 3.09 ANKRD43 3.00 ELAC2 2.94 MMP11 2.87 PSMD10 3.09 RARA 3.00 AA807922 2.93 MTMR2 2.86 PGBD2 3.09 ABCA12 3.00 ZCSL2 2.93 PNPLA4 2.86 CAMP 3.08 C14orf179 3.00 BNIP2 2.93 CCDC32 2.86 SMCR5 3.08 THC2274391 3.00 STARD5 2.93 A_24_P264549 2.85 TRAPPC4 3.08 SLC31A2 3.00 GUSBL2 2.92 ZC3H12A 2.85 ENST00000267857 3.08 THC2306096 3.00 RAGE 2.92 ENST00000379534 2.85 C20orf55 3.08 FLJ20294 3.00 GFM2 2.92 CLK3 2.85 ZNF407 3.08 AK022339 2.99 A_24_P204015 2.92 A_24_P927553 2.85 TP73L 3.08 ZC3H14 2.99 WDR61 2.92 SRF 2.85 THC2320516 3.07 MYO1E 2.99 PNMA3 2.92 SLFN12 2.85 RIC3 3.07 INHBC 2.99 CHURC1 2.92 CENTB2 2.84 NINJ1 3.07 HPS1 2.99 FLJ30064 2.92 BM955917 2.84

117

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) RPS6KA5 2.60 BCORL1 2.55 ENST00000356572 2.49 CRELD1 2.44 A_24_P480722 2.60 WNK1 2.55 C6orf162 2.49 PSPH 2.44 ENST00000301807 2.59 EIF1B 2.54 THC2434618 2.49 C12orf42 2.44 ZFHX2 2.59 A_24_P290114 2.54 A_32_P31206 2.49 LOC440396 2.44 PARC 2.59 KIAA1826 2.54 RPL26L1 2.49 PKD2 2.44 A_24_P15502 2.59 KIAA0226 2.54 PARP14 2.49 THC2364821 2.43 WDR60 2.59 UBE4B 2.54 THC2440027 2.49 RBM18 2.43 LOC644701 2.59 SPFH2 2.54 ID2 2.49 ENST00000316634 2.43 C18orf25 2.59 ZCCHC14 2.54 MED28 2.48 LOC441876 2.43 A_23_P208579 2.59 C3orf54 2.54 BG952851 2.48 A_32_P69379 2.43 CA438802 2.59 ENST00000378953 2.54 FLJ22222 2.48 SRCRB4D 2.43 PRKX 2.59 MT1H 2.54 OXNAD1 2.48 CYB561 2.43 PCYT1A 2.59 HSPC268 2.53 C21orf91 2.48 FAM49B 2.43 A_24_P367399 2.59 FAM123B 2.53 A_24_P932046 2.48 DBF4B 2.43 GPR84 2.59 TCF2 2.53 PSTPIP1 2.48 N75321 2.43 UNQ1887 2.59 ENST00000374929 2.53 ACCN1 2.48 TOM1L2 2.43 MYH14 2.59 RIPK5 2.53 THC2274051 2.48 C1orf121 2.43 FKSG24 2.58 CKAP2 2.53 ENST00000324677 2.48 AK092942 2.43 LHX4 2.58 TMEM57 2.53 AI263083 2.48 ENST00000290390 2.42 SIDT2 2.58 VMD2L2 2.53 ZNF613 2.47 RAB15 2.42 DZIP3 2.58 CR620804 2.53 SGCB 2.47 C12orf51 2.42 AL390181 2.58 FOXC1 2.53 A_24_P144625 2.47 CR605438 2.42 EPOR 2.58 ZNF138 2.53 THC2437430 2.47 ZNF688 2.42 A_24_P41662 2.58 AF116619 2.53 AF086288 2.47 BC047753 2.42 BC046172 2.58 TSGA10 2.53 GGTL3 2.47 PLEKHG6 2.42 IL23R 2.58 MGC16385 2.53 APH1B 2.47 ADIPOQ 2.42 ELAC1 2.58 AK095600 2.53 BX115105 2.47 DIP2B 2.42 A_24_P533142 2.57 RSHL2 2.53 RP4-756G23.1 2.47 A_32_P54260 2.42 THC2343880 2.57 NOS3 2.52 FLJ36031 2.47 ALAS1 2.42 CARD6 2.57 THC2277571 2.52 THC2439773 2.47 A_24_P272735 2.41 TRIM6 2.57 LOC645919 2.52 NDUFB6 2.46 POLK 2.41 STOML1 2.57 C20orf112 2.52 WDR1 2.46 KRTAP6-3 2.41 SMCR7 2.57 MLLT6 2.52 LOC339804 2.46 TAF12 2.41 ENST00000309829 2.57 MAP1D 2.52 ENST00000329627 2.46 P2RY10 2.41 A_32_P109645 2.57 HELZ 2.52 CCDC97 2.46 ENST00000322831 2.41 MAPKBP1 2.57 SCAMP5 2.52 AA873311 2.46 BC011998 2.41 A_24_P913620 2.57 THC2310680 2.52 IBRDC3 2.46 PDDC1 2.41 ZNF527 2.56 LOC647219 2.52 A_32_P134679 2.46 ANKRD11 2.41 BX448200 2.56 A_24_P25063 2.51 AK123446 2.46 KIAA1908 2.41 ENST00000317633 2.56 COMMD9 2.51 DYNLL2 2.46 KIAA1383 2.41 SRP46 2.56 A_32_P164061 2.51 HARS 2.46 PIK3R3 2.41 VPS18 2.56 C6orf157 2.51 A_32_P8971 2.46 KIAA1429 2.41 CENTG3 2.56 FMNL3 2.51 ZNF549 2.46 HRASLS2 2.41 ELP3 2.56 KREMEN1 2.51 CLEC4M 2.46 BMP2K 2.41 MRPL10 2.56 PARS2 2.51 BQ000122 2.46 CMTM7 2.40 GALNT11 2.56 A_23_P329062 2.51 MGC15523 2.46 IL27RA 2.40 GHRL 2.56 SLC43A2 2.50 RNF14 2.45 ITGB1 2.40 HIF1A 2.56 ZNF286 2.50 AGER 2.45 AK055981 2.40 AK096685 2.56 BE672985 2.50 PLEKHG1 2.45 AK074776 2.40 AK127258 2.55 CTSL 2.50 ANKRD17 2.45 THC2341060 2.40 NGFRAP1L1 2.55 DKFZp547C195 2.50 MKNK1 2.45 SGCA 2.40 VHL 2.55 AD7C-NTP 2.50 ARL17P1 2.45 CRIPAK 2.40 AK090827 2.55 BX537816 2.50 LOC541472 2.45 PEX14 2.40 SLC25A14 2.55 IGLL1 2.50 BC036431 2.45 ZNF8 2.40 A_32_P225916 2.55 ARMC9 2.50 KCNC3 2.45 MRPL39 2.40 RAB5A 2.55 FCGBP 2.50 C15orf20 2.45 B4GALT3 2.40 CCDC57 2.55 C9orf95 2.49 SLC16A8 2.45 MAK10 2.39 A_24_P127543 2.55 A_24_P850187 2.49 CPT2 2.45 NALP1 2.39 DNMT3A 2.55 THC2442550 2.49 FANCC 2.45 AK097080 2.39

118

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) SIDT1 2.39 AK128192 2.35 GPKOW 2.30 UBE2B 2.26 MFSD1 2.39 15E1.2 2.35 THC2308747 2.30 SEMA7A 2.26 FNDC3B 2.39 FOXO1A 2.35 ENST00000379734 2.30 EEA1 2.26 A_24_P698816 2.39 SLC6A4 2.35 DDX28 2.30 A_24_P15083 2.26 CHCHD5 2.39 AF132973 2.35 LMBR1 2.30 BC017848 2.26 ATG7 2.39 TM9SF1 2.35 HAGH 2.30 THC2266506 2.26 BC000604 2.39 BI759100 2.35 MLLT7 2.30 CDADC1 2.26 ZCCHC3 2.39 BAZ2B 2.35 ZNF226 2.30 ENST00000354743 2.26 XCL1 2.39 THC2265980 2.35 SEC61A2 2.30 KIAA1279 2.26 CMTM3 2.39 ENST00000342275 2.34 SCC-112 2.30 NOTCH2NL 2.26 BC015962 2.39 ZNF33A 2.34 ATP6V1H 2.30 AW874684 2.26 HDAC4 2.39 THC2336549 2.34 ENST00000381108 2.30 TMEM85 2.26 ENST00000381929 2.39 A_24_P681266 2.34 LOC205251 2.30 HAAO 2.26 C1orf117 2.39 C19orf44 2.34 THC2366887 2.29 TBKBP1 2.26 C15orf23 2.39 BC031316 2.34 ZZZ3 2.29 BC073976 2.26 A_24_P101879 2.39 ZBTB7B 2.34 ENST00000380357 2.29 LOC440354 2.25 ADAMTS13 2.39 AI887037 2.34 THC2318533 2.29 ENST00000375672 2.25 IL21R 2.39 METTL8 2.34 BX421789 2.29 FGD3 2.25 ENST00000317868 2.38 LZTR2 2.34 tcag7.1239 2.29 ICT1 2.25 ZNF329 2.38 RNF141 2.34 NUP43 2.29 AK125129 2.25 AARSL 2.38 RDHE2 2.34 SPSB4 2.29 BC030102 2.25 A_24_P290068 2.38 THC2337763 2.33 BC002811 2.29 B3GALT4 2.25 OTUD1 2.38 PXMP4 2.33 PAOX 2.29 VAMP3 2.25 A_23_P217187 2.38 ALG1 2.33 THC2317149 2.29 CUL5 2.25 PORCN 2.38 FBXO31 2.33 PITPNA 2.29 C7 2.25 ZNFN1A4 2.38 THRA 2.33 ARL6IP 2.29 A_24_P383834 2.25 HDDC2 2.38 PML 2.33 MTERFD2 2.28 ENST00000376068 2.25 CXX1 2.38 GNG13 2.33 A_32_P24431 2.28 PLEK 2.25 DCTN6 2.38 AK098185 2.33 C13orf1 2.28 ZDHHC24 2.25 SULT1A2 2.38 C9orf97 2.33 APOL2 2.28 CAMLG 2.25 C7orf27 2.38 THC2363295 2.33 SLC34A1 2.28 GNPTAB 2.25 PIGT 2.38 FLJ32255 2.32 THC2399982 2.28 PCDHGA8 2.25 A_24_P332263 2.38 GAS2L1 2.32 THC2407694 2.28 ENST00000297423 2.25 UBE3B 2.38 GRINL1A 2.32 LIN7A 2.28 THC2372128 2.25 SLC9A3R2 2.38 PRCP 2.32 NEXN 2.28 SMUG1 2.25 BOLA3 2.38 NMT1 2.32 MGC19604 2.28 SILV 2.25 DIAPH3 2.38 CR992331 2.32 SOCS1 2.28 AK023660 2.24 ENST00000358916 2.37 CGRRF1 2.32 RHPN1 2.28 PFDN4 2.24 DPYSL3 2.37 GLIPR1 2.32 ST3GAL2 2.28 ZNF587 2.24 SNAPC5 2.37 THC2406192 2.32 BF939434 2.28 BG284526 2.24 THC2342491 2.37 A_32_P8857 2.32 A_24_P843552 2.28 LRRC49 2.24 ZNF597 2.37 ENST00000311528 2.32 CD58 2.28 WFDC3 2.24 C14orf131 2.37 BM975266 2.32 ENST00000335142 2.28 MARVELD2 2.24 MT1G 2.37 DMPK 2.31 FAM21C 2.28 A_23_P108534 2.24 SPSB1 2.37 A_24_P358205 2.31 C14orf147 2.28 CD80 2.24 AK057088 2.37 THC2343253 2.31 AF116719 2.27 SMAD3 2.24 DEPDC7 2.37 GUCY2C 2.31 LIN7B 2.27 RASGRF1 2.24 NINJ2 2.36 PYGO2 2.31 ZMYM1 2.27 C10orf32 2.23 MON2 2.36 LY6G5C 2.31 FAM60A 2.27 ENST00000349637 2.23 CYHR1 2.36 A_24_P196024 2.31 PRKCI 2.27 THC2365798 2.23 C18orf1 2.36 PSEN2 2.31 PPID 2.27 ENST00000366971 2.23 SLC41A3 2.36 ZNF284 2.31 SLC9A6 2.27 CMTM6 2.23 A_24_P101211 2.36 CRY2 2.31 AK055501 2.27 ENST00000330044 2.23 C6orf70 2.36 GATS 2.31 ATG16L2 2.27 RFX3 2.23 KIAA0494 2.36 CASP8 2.31 AW977527 2.27 NCF2 2.23 SYK 2.36 ABHD4 2.31 KIAA1632 2.27 THC2426662 2.23 CXCL3 2.36 LOC440905 2.31 THC2275950 2.27 RFXDC2 2.23 USP20 2.35 CASKIN1 2.30 SLA/LP 2.27 BM982926 2.23 ZNF77 2.35 NUDT14 2.30 WDR48 2.27 KIAA1443 2.23

119

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE UP-REGULATED GENES (3430) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) FAIM3 2.23 ZNF211 2.19 KLHL25 2.15 COQ5 2.11 ATP6V1A 2.23 ENST00000360514 2.19 BTN2A1 2.15 CHCHD7 2.11 MOCOS 2.23 A_24_P118382 2.19 TNFSF9 2.15 ENST00000333546 2.11 UGT2B17 2.22 A_24_P884915 2.19 PRKCH 2.15 BX538248 2.11 ZNF161 2.22 TMEM68 2.19 AW948903 2.15 BX090181 2.11 GBP3 2.22 SIRPA 2.19 THC2305638 2.15 DNHD1 2.11 BC080577 2.22 ENST00000257951 2.19 GRPEL2 2.15 AK130207 2.11 SRP19 2.22 ENST00000327026 2.18 LRP10 2.15 USP6NL 2.11 TBXAS1 2.22 VPS41 2.18 GNB3 2.15 SECTM1 2.11 LDHC 2.22 GIMAP8 2.18 MGC3207 2.15 A_24_P670147 2.11 HLA-DOA 2.22 C17orf37 2.18 A_23_P30411 2.14 EMILIN3 2.11 FBXO28 2.22 MGC40499 2.18 CCDC56 2.14 ENST00000367545 2.11 THTPA 2.22 PFTK1 2.18 CPT1A 2.14 ATF7 2.10 GLYCTK 2.22 THC2249577 2.18 KNTC2 2.14 ENST00000329367 2.10 SP1 2.22 PRO2900 2.18 RCBTB1 2.14 CNOT7 2.10 F8 2.22 L3MBTL 2.18 PPAPDC1B 2.14 P18SRP 2.10 TCF15 2.22 CD86 2.18 NOD3 2.14 C14orf43 2.10 NRL 2.22 DLG2 2.18 PIGB 2.14 BX352604 2.10 AF289590 2.22 IRX5 2.18 THC2374943 2.14 THC2395899 2.10 A_24_P631948 2.21 MYST4 2.18 THC2406285 2.14 BC051368 2.10 IL28RA 2.21 THC2269516 2.18 MTHFD1L 2.14 PNKD 2.10 KIAA0256 2.21 C6orf85 2.18 AK130530 2.14 C6orf35 2.10 LOC400509 2.21 C19orf12 2.18 IFT20 2.14 AK026485 2.10 MT1E 2.21 A_32_P80295 2.17 C8orf47 2.14 VPS37B 2.10 ZNF690 2.21 CTNND1 2.17 ENST00000374390 2.14 CYP1B1 2.10 C14orf65 2.21 KIAA0141 2.17 BC028022 2.14 HPRT1 2.10 GOLPH3L 2.21 A_24_P400751 2.17 ZNF701 2.14 C14orf167 2.10 SLC25A20 2.21 HOXB3 2.17 MAPK13 2.13 DTX3 2.10 ORC6L 2.21 A_24_P761386 2.17 CASP10 2.13 CD99L2 2.10 UCRC 2.21 AK055915 2.17 SEDLP 2.13 THC2400529 2.10 DKFZp667E0512 2.21 A_32_P131870 2.17 TSPYL5 2.13 AA479896 2.09 KIAA1856 2.21 GP9 2.17 UFD1L 2.13 CDCA2 2.09 RLN2 2.21 KLK1 2.17 NKG7 2.13 SMARCD1 2.09 LOC400236 2.21 THC2284657 2.17 GCC2 2.13 GNG11 2.09 THC2434173 2.21 BC014218 2.17 CSTA 2.13 WIBG 2.09 IMPAD1 2.20 CHRNB2 2.17 ENST00000222396 2.13 WHSC1L1 2.09 RNF135 2.20 THC2375545 2.17 CTNNBIP1 2.13 IER5L 2.09 ENST00000377548 2.20 C20orf74 2.16 AA627222 2.13 AK098749 2.09 TRIM59 2.20 GIMAP1 2.16 C1orf124 2.13 MGC33556 2.09 TIGD6 2.20 A_24_P33217 2.16 UBE2F 2.13 FAM104A 2.09 GGA3 2.20 WDR66 2.16 NUDT1 2.12 LOC283551 2.09 NDUFA5 2.20 SENP8 2.16 BTN2A2 2.12 TNS3 2.09 TINF2 2.20 BG209623 2.16 TMEM2 2.12 ACSS1 2.09 ENST00000303979 2.20 CCAR1 2.16 S100A13 2.12 TTC7A 2.09 SCGB3A2 2.20 THC2302971 2.16 THC2382871 2.12 BC041926 2.09 THC2433033 2.20 ENST00000305749 2.16 BX375060 2.12 FAM53C 2.08 NAPG 2.20 TRAF7 2.16 LOC349114 2.12 BQ614035 2.08 C3orf1 2.19 ANXA2 2.16 THC2395355 2.12 CBLL1 2.08 DSTN 2.19 MGC34796 2.16 MAPK14 2.12 TRAF3 2.08 ZNF33B 2.19 R3HDM2 2.16 CCR1 2.12 THC2375021 2.08 AURKA 2.19 A_24_P358305 2.16 AV757313 2.12 C7orf29 2.08 MAP2K4 2.19 BTN2A3 2.16 RPL23AP7 2.12 LOC222699 2.08 RASAL1 2.19 RUNDC2A 2.15 AK126405 2.11 PARN 2.08 C9orf6 2.19 COX18 2.15 TMUB2 2.11 AW797858 2.08 FASTKD2 2.19 IL13RA1 2.15 BCAS4 2.11 BTG2 2.08 C20orf23 2.19 FECH 2.15 PDPK1 2.11 SLC24A1 2.08 BX104999 2.19 TAS1R1 2.15 SUV420H1 2.11 MT1M 2.08 ATF6 2.19 ENST00000326261 2.15 ZNF133 2.11 RAD51C 2.07 ZBTB8 2.19 TMEM50B 2.15 BC017972 2.11 DDR1 2.06

120

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2728) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) UNG -8.29 SLC1A1 -4.51 NOLC1 -4.08 FUT11 -3.84 C1QTNF5 -7.17 TJP2 -4.50 CAMK2D -4.08 UNC5CL -3.84 THC2357547 -6.59 CDK2 -4.50 PDXP -4.06 BM932296 -3.83 C12orf34 -6.53 ZNF395 -4.48 ENST00000315293 -4.06 C3orf21 -3.83 DTL -6.20 NBL1 -4.48 HM13 -4.06 ENST00000382990 -3.82 CDC6 -6.18 DMC1 -4.48 NAALADL1 -4.05 ENST00000332107 -3.82 KLHL23 -6.11 EPN1 -4.47 BM984383 -4.04 PHLPP -3.81 ENST00000375256 -6.03 ENST00000369158 -4.47 MAPK9 -4.04 PYCR1 -3.81 TEAD4 -6.02 C16orf55 -4.47 TIGD7 -4.04 AP2A1 -3.80 A_24_P195400 -5.77 PUM2 -4.45 ST3GAL1 -4.03 CDT1 -3.80 CCNE2 -5.70 BC064349 -4.45 ARHGEF1 -4.03 AHNAK -3.80 FKBP5 -5.65 SET -4.45 RNASET2 -4.03 CTSC -3.79 NR1D2 -5.64 GPR146 -4.44 METAP1 -4.02 HCFC1R1 -3.78 PAQR4 -5.64 ENO2 -4.42 AKT1 -4.02 KIAA1008 -3.78 MCM4 -5.47 SLC43A1 -4.41 TRIM65 -4.02 PBEF1 -3.78 CDCA7 -5.46 SCML1 -4.40 MGC13017 -4.01 LOC442013 -3.77 CHAF1B -5.41 PSMD11 -4.40 HPCA -4.01 AK094415 -3.77 ZADH2 -5.30 THC2404842 -4.38 ASB13 -4.00 A_23_P2032 -3.76 FEN1 -5.25 UHRF2 -4.36 SLC16A1 -4.00 SERPINH1 -3.76 FGFRL1 -5.14 THC2374166 -4.35 YEATS4 -4.00 ORC1L -3.76 SLC1A5 -5.06 TNFRSF12A -4.34 RFC4 -3.99 TCF3 -3.76 RHOBTB2 -5.05 RHEB -4.33 KIAA0319L -3.99 TNPO1 -3.76 NCKIPSD -5.03 AK123655 -4.32 ENST00000377515 -3.99 BC022233 -3.75 GPR155 -5.02 LRRC20 -4.32 CB853344 -3.98 NOL5A -3.75 THC2309258 -5.01 LIMD1 -4.31 KIAA1109 -3.98 MBD3 -3.75 POLD1 -4.99 C17orf69 -4.29 EXO1 -3.98 DDN -3.75 ETS2 -4.99 MLLT4 -4.29 A_24_P810074 -3.98 ATAD2 -3.74 CDCA7L -4.97 PGRMC2 -4.25 INADL -3.98 THC2431161 -3.74 GPR89A -4.96 ENST00000311630 -4.24 A_23_P31563 -3.98 A_24_P594094 -3.74 PSIP1 -4.94 ENST00000312785 -4.24 ZNF414 -3.97 TMEM16F -3.74 FKBP4 -4.90 ARL6IP6 -4.23 HSF2 -3.97 FLJ40542 -3.73 GPR30 -4.86 MTAP -4.23 A_32_P219704 -3.95 TM9SF4 -3.73 EIF2S2 -4.86 POLD3 -4.22 ANKRD28 -3.95 APC -3.73 XPOT -4.81 COPA -4.21 ENST00000308269 -3.95 SMC3 -3.72 TFDP3 -4.81 LOC644063 -4.21 ENST00000334464 -3.94 GSPT1 -3.72 GINS2 -4.81 FGFR1 -4.21 MSH2 -3.94 ENST00000370290 -3.72 CDC25A -4.75 MYOHD1 -4.19 THC2276742 -3.93 WNT6 -3.72 GNG7 -4.74 C10orf119 -4.18 FLJ39779 -3.93 LOC388114 -3.72 NUP205 -4.72 ING2 -4.17 MINK1 -3.92 A_24_P200962 -3.72 STS -4.71 RNPS1 -4.17 EXOC6 -3.92 C10orf137 -3.71 TUBGCP3 -4.70 TMEM106C -4.16 TMEM44 -3.92 EPB41L2 -3.71 ACY1L2 -4.70 PIGA -4.16 PARP11 -3.91 AGPAT2 -3.71 CD96 -4.68 OXCT2 -4.16 SPIN -3.91 ABCA2 -3.71 BC079833 -4.63 MCM10 -4.16 ENST00000367740 -3.90 SLC7A5 -3.71 GLOXD1 -4.62 A_24_P358302 -4.14 GRAP -3.90 PRPS2 -3.70 PPP5C -4.62 WDR76 -4.14 ENST00000369731 -3.90 GPR157 -3.70 LOC643431 -4.62 RPP25 -4.14 C14orf43 -3.89 IDI1 -3.70 RCC1 -4.61 ZMYND19 -4.14 DEK -3.88 TAGLN2 -3.69 CR613972 -4.61 MGC40405 -4.13 PPAT -3.87 YARS -3.69 C5orf25 -4.60 FLJ30596 -4.13 ADCK2 -3.87 VLDLR -3.69 BARD1 -4.60 A_24_P375962 -4.12 WWOX -3.86 ENST00000318251 -3.68 KIAA1715 -4.58 AYTL2 -4.12 SIAH1 -3.86 SIX5 -3.68 UHRF1 -4.58 EHMT2 -4.12 BE003490 -3.86 ATAD1 -3.68 A_24_P221485 -4.55 CABLES1 -4.11 RAB1B -3.85 TMTC4 -3.67 LARP1 -4.53 GART -4.11 ING1 -3.85 SLAMF6 -3.67 RBM21 -4.53 ELMO2 -4.11 BC036361 -3.85 OPN3 -3.67 SYMPK -4.53 MAZ -4.09 PPME1 -3.84 RASSF1 -3.67 BC031876 -4.53 VAMP2 -4.09 CHDH -3.84 USP28 -3.66 CR625518 -4.51 ENST00000332534 -4.09 SMAP1 -3.84 C20orf59 -3.66

121

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2728) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) DTX1 -3.66 ZNF687 -3.53 CENPQ -3.43 TSC22D2 -3.34 THC2320434 -3.66 RARG -3.53 HK2 -3.43 ANKRD41 -3.34 SUV39H2 -3.66 A_24_P233560 -3.53 MESP1 -3.43 KLHDC3 -3.34 PES1 -3.66 ORC6L -3.53 A_24_P383751 -3.43 ZRANB1 -3.34 THC2343426 -3.66 SYTL3 -3.53 CIRH1A -3.43 FAM35A -3.34 BRD2 -3.66 AF445027 -3.53 AK057981 -3.43 MMP25 -3.34 GDI2 -3.66 RNF167 -3.53 AK092888 -3.42 ITGB1BP1 -3.33 P2RY11 -3.65 ANP32B -3.53 AKAP8L -3.42 GEMIN5 -3.33 KCTD20 -3.65 DEPDC4 -3.52 MGC4562 -3.42 SMC5 -3.33 IMPDH1 -3.65 CCDC71 -3.52 TFDP1 -3.42 DBP -3.33 BXDC2 -3.64 KLHL22 -3.52 LOC200420 -3.42 FASN -3.33 SREBF1 -3.64 KDELC2 -3.52 TMED2 -3.42 INSIG2 -3.33 ZNF473 -3.64 LRRC42 -3.52 TAF1L -3.42 THC2376568 -3.33 ENST00000359244 -3.64 SMARCB1 -3.52 DA234975 -3.42 THC2428103 -3.33 D89937 -3.64 ATXN7L3 -3.52 ZNF514 -3.42 STARD3 -3.33 ZNF146 -3.63 ZNF525 -3.51 BC047380 -3.42 PIGX -3.33 GTF2I -3.63 AFG3L2 -3.51 BE467780 -3.41 FZD1 -3.33 CHKA -3.63 KIAA1509 -3.51 ZBED3 -3.41 AF086536 -3.32 RTF1 -3.63 DDX11 -3.51 MGC13005 -3.40 NUPL1 -3.32 GRPEL1 -3.63 SPEN -3.51 A_24_P247616 -3.40 USP37 -3.32 UAP1 -3.63 FARSLB -3.51 CHD8 -3.40 PHF20 -3.32 NUDCD3 -3.63 CX164944 -3.51 A_24_P539275 -3.40 AK057591 -3.32 CD300C -3.62 FBXL10 -3.51 LOC339344 -3.39 AK091337 -3.32 DDX55 -3.62 PTGES3 -3.51 THC2357608 -3.39 PTP4A2 -3.32 PTK2B -3.61 STK24 -3.50 SIAH2 -3.39 CHIC1 -3.32 AK023774 -3.61 GLUD2 -3.50 STX7 -3.39 OTUD4 -3.31 MTHFD1L -3.61 ENST00000344035 -3.50 UPP1 -3.39 UBQLN4 -3.31 A_24_P50281 -3.60 09/01/2006 -3.50 THC2268736 -3.39 AF132203 -3.31 SETD1A -3.60 CREBBP -3.49 C1orf63 -3.39 TXNDC4 -3.31 A_24_P221601 -3.60 DLG3 -3.49 FLJ25715 -3.38 TCTE3 -3.31 CUTC -3.60 LOC375133 -3.48 A_24_P273014 -3.38 A_24_P92823 -3.31 PHF17 -3.60 SIPA1 -3.48 YWHAE -3.38 ENST00000383518 -3.31 AK095904 -3.60 LOC643668 -3.48 A_24_P714707 -3.38 FOXJ2 -3.30 A_24_P118721 -3.59 UBQLN1 -3.48 MRPL19 -3.38 POU5F1 -3.30 MVK -3.59 A_24_P853302 -3.48 KHSRP -3.38 TMEM121 -3.30 ING5 -3.58 PLCXD1 -3.48 A_24_P349648 -3.37 YES1 -3.30 UBA2 -3.58 TMEM81 -3.48 A_24_P203976 -3.37 GGA2 -3.30 PAXIP1 -3.58 HERC2 -3.48 C20orf20 -3.37 DCPS -3.30 MAPK14 -3.58 MLL3 -3.48 ENST00000359653 -3.37 PFKFB3 -3.30 POLE -3.58 CLSPN -3.48 KCNQ5 -3.37 NAT13 -3.29 BE816002 -3.58 A_24_P401124 -3.48 ENST00000367142 -3.37 NUDT9P1 -3.29 COX19 -3.57 CSNK2A1 -3.48 N-PAC -3.36 HIVEP2 -3.29 D21S2056E -3.57 THC2405842 -3.47 A_32_P114268 -3.36 ENST00000357180 -3.29 DEGS1 -3.57 PPP1R7 -3.47 A_24_P170309 -3.36 PABPC4 -3.29 CCDC85B -3.57 TNKS1BP1 -3.47 PLK4 -3.36 THC2363476 -3.29 SPFH1 -3.57 AF090926 -3.46 AHI1 -3.36 PRKCSH -3.29 SNRPB -3.57 FLJ12529 -3.46 C10orf9 -3.36 ENST00000382579 -3.29 ATP6V1C1 -3.57 GMPS -3.46 SNX9 -3.36 POLR3K -3.28 PLXNC1 -3.57 AES -3.46 ELF4 -3.36 CHRAC1 -3.28 THC2407148 -3.56 CCDC109A -3.46 GALNTL4 -3.35 SLC39A8 -3.28 A_24_P692600 -3.55 KLHL17 -3.46 FEM1C -3.35 XBP1 -3.28 SLC7A11 -3.55 THC2313920 -3.45 CR612518 -3.35 BHLHB3 -3.28 LY9 -3.55 TLOC1 -3.45 SFRS1 -3.35 CELSR3 -3.28 BTBD14B -3.54 PUS7 -3.45 THC2305336 -3.35 VCP -3.28 RFC1 -3.54 THC2400533 -3.44 PKP4 -3.35 DENR -3.28 AK126814 -3.54 HNRPD -3.44 SLC19A1 -3.35 UAP1L1 -3.28 SGK3 -3.54 GANAB -3.44 ST6GALNAC6 -3.35 DDX3X -3.28 SYNJ2 -3.54 AF119906 -3.44 CAMK1D -3.35 LARP2 -3.27 OGG1 -3.54 ZFAND2B -3.43 ZNF397 -3.35 GON4L -3.27

122

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2728) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) A_24_P835943 -3.27 KIAA1815 -3.19 A_32_P78488 -3.12 ERO1L -3.05 C12orf41 -3.27 NEDD4L -3.18 PAK2 -3.12 C14orf24 -3.05 NOB1 -3.27 LOC388796 -3.18 ARFRP1 -3.12 LOC441320 -3.05 CR603272 -3.27 BCL2L11 -3.18 TOMM40 -3.12 INPP5F -3.05 ZFP161 -3.27 LOC152719 -3.18 A_24_P101742 -3.11 PGEA1 -3.05 AK021546 -3.27 LYAR -3.18 PER1 -3.11 AHSA1 -3.04 CNOT3 -3.27 A_24_P625898 -3.18 KDELR1 -3.11 TPCN1 -3.04 PARP16 -3.26 CCNG2 -3.18 FLNA -3.11 ENO1B -3.04 ASNA1 -3.26 CHRNA5 -3.18 C19orf22 -3.11 TUBA1 -3.04 CR616772 -3.25 MLLT6 -3.17 GNPNAT1 -3.11 SOX12 -3.04 CR609948 -3.25 RP2 -3.17 MRPL17 -3.11 UBL4A -3.04 ATF4 -3.25 A_32_P55414 -3.17 BF869497 -3.11 POLR1A -3.03 SLC20A1 -3.25 SEPN1 -3.17 HABP4 -3.11 AK021629 -3.03 A_24_P332721 -3.25 LSM12 -3.17 LAT2 -3.11 A_24_P24453 -3.03 AK054718 -3.25 NP285481 -3.17 ENSA -3.10 THC2306239 -3.03 NGLY1 -3.25 TUBE1 -3.17 DCC1 -3.10 EEF2K -3.03 SMA4 -3.25 AKT1S1 -3.16 LOC221272 -3.10 C9orf41 -3.03 A_24_P796274 -3.25 PRPF18 -3.16 HMGN4 -3.10 WDHD1 -3.03 ZBTB6 -3.25 PIP5K1A -3.16 THC2375651 -3.10 CCDC93 -3.03 NCL -3.24 ENST00000269142 -3.16 ST6GAL1 -3.10 THC2407189 -3.03 SLC31A1 -3.24 ESCO1 -3.16 CENPB -3.10 A_24_P255865 -3.02 CACNB3 -3.24 C6orf89 -3.16 CAMKK1 -3.10 PRKCBP1 -3.02 A_24_P409816 -3.24 KIAA2013 -3.16 DNMT1 -3.10 RBM23 -3.02 ENST00000265341 -3.24 AK097332 -3.16 TCF20 -3.10 RP11-262H14.4 -3.02 AHCTF1 -3.24 SKP2 -3.16 AARS -3.10 THC2409354 -3.02 ENST00000344771 -3.24 LOC653458 -3.16 CGI-09 -3.10 PMS1 -3.02 CHM -3.23 TBL2 -3.16 ENST00000380635 -3.10 A_24_P213336 -3.02 RPL7L1 -3.23 KIAA1285 -3.16 P4HA1 -3.10 STK35 -3.02 JMJD1C -3.23 MRPS2 -3.15 MLSTD1 -3.10 IL18R1 -3.02 TP73 -3.23 MAD2L1 -3.15 AF116678 -3.10 NOC2L -3.02 THC2344809 -3.22 KPNB1 -3.15 ADAM9 -3.10 AF086045 -3.02 CAV1 -3.22 BTN3A1 -3.15 LOC133619 -3.09 WDR79 -3.01 FADS2 -3.22 RFC2 -3.15 LOC339123 -3.09 CHMP7 -3.01 SC4MOL -3.22 BX094072 -3.15 TEC -3.09 PPP1R3E -3.01 THC2250386 -3.22 ABCC4 -3.15 UBE2MP1 -3.09 PTGES2 -3.01 CENTA1 -3.22 ORM1 -3.15 GINS3 -3.09 ARHGDIA -3.01 SCMH1 -3.21 L07392 -3.15 FAM29A -3.09 HSD17B6 -3.01 CAMKK2 -3.21 A_32_P177300 -3.15 SCD -3.09 MGC23280 -3.01 MSN -3.21 BC045174 -3.14 A_24_P803809 -3.09 C22orf32 -3.01 CENTG3 -3.21 ZBTB2 -3.14 ZNF566 -3.09 ZNF569 -3.00 TPD52 -3.21 CRSP6 -3.14 ENST00000246083 -3.09 THAP6 -3.00 NDST1 -3.21 AK021676 -3.14 C9orf74 -3.08 TMEM93 -3.00 PPAN -3.21 HDAC2 -3.14 BAT2 -3.08 C9orf32 -3.00 TTC19 -3.21 SLC16A3 -3.14 CR604908 -3.08 FLJ10154 -3.00 C8orf33 -3.20 GLT25D1 -3.14 VPS54 -3.07 BF965065 -3.00 EGLN1 -3.20 NUP155 -3.14 A_24_P921801 -3.07 AA843546 -3.00 PTCD2 -3.20 GPIAP1 -3.13 TNPO3 -3.07 CCDC21 -3.00 NUFIP1 -3.20 BX346853 -3.13 A_24_P418498 -3.07 THC2270231 -3.00 FXR2 -3.20 ARHGEF12 -3.13 BMS1L -3.07 C10orf95 -3.00 VEGFB -3.20 M6PR -3.13 MGC4655 -3.06 ENST00000341591 -3.00 AMPD2 -3.20 BC031940 -3.13 A_24_P24806 -3.06 PLCH2 -3.00 GCN5L2 -3.19 ZNHIT4 -3.13 ACSL1 -3.06 ETV4 -3.00 KIAA0664 -3.19 BAG5 -3.12 A_24_P771278 -3.06 THAP7 -3.00 CTGLF1 -3.19 CLK2 -3.12 GMNN -3.06 TOM1 -3.00 KIAA0372 -3.19 NANP -3.12 TROVE2 -3.06 PDK1 -2.99 FOXJ3 -3.19 A_24_P170283 -3.12 PWP2H -3.06 THC2338942 -2.99 A_24_P530900 -3.19 A_23_P251196 -3.12 CEBPG -3.05 ZNF23 -2.99 VPS26B -3.19 EIF4EBP1 -3.12 FUT7 -3.05 STIP1 -2.99 MASTL -3.19 ZC3HAV1L -3.12 GEMIN4 -3.05 A_32_P108420 -2.99

123

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2728) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) SIT1 -2.99 APOL1 -2.94 IER3IP1 -2.89 XTP3TPA -2.85 ADM -2.99 A_24_P75708 -2.94 IVD -2.89 A_23_P300563 -2.85 ELK1 -2.99 ILK -2.94 MANEAL -2.89 SLFN11 -2.85 TMEM86B -2.99 SLC39A3 -2.94 FLJ22639 -2.89 ENST00000373219 -2.84 POU3F1 -2.99 A_24_P846755 -2.94 N4BP3 -2.89 GRPEL2 -2.84 KCNH3 -2.99 THC2337923 -2.94 MBD1 -2.89 BG259069 -2.84 CR596712 -2.99 HES5 -2.94 TMEM39B -2.89 THC2273298 -2.84 PDDC1 -2.99 GMPPB -2.94 SSPN -2.89 FLJ20245 -2.84 TPD52L2 -2.99 MFN1 -2.94 ANP32D -2.88 A_24_P470809 -2.84 PRIM1 -2.99 DNAJB2 -2.94 SMN2 -2.88 ENST00000378887 -2.84 RAB5A -2.99 C14orf58 -2.93 LUZP5 -2.88 RNF122 -2.83 SELI -2.99 DHCR24 -2.93 A_24_P341489 -2.88 DONSON -2.83 DNAL4 -2.98 BAT1 -2.93 RBM38 -2.88 WHSC1 -2.83 A_24_P565908 -2.98 SETDB1 -2.93 PCTK2 -2.88 CXorf9 -2.83 FLJ40722 -2.98 HNRPDL -2.93 CDKN2C -2.88 BC033052 -2.83 C19orf28 -2.98 A_24_P169574 -2.93 ENST00000335459 -2.88 SUV420H2 -2.83 A_24_P307424 -2.98 A_24_P255845 -2.93 ESAM -2.88 ACVR2B -2.83 MOBK1B -2.98 SLC27A4 -2.93 GNPDA1 -2.88 A_24_P324214 -2.83 ALDH1L2 -2.98 SPAG9 -2.93 BECN1 -2.87 CUL1 -2.83 ENST00000328046 -2.98 FLJ36874 -2.93 CCDC28B -2.87 A_32_P6274 -2.83 THC2373624 -2.98 LOC646791 -2.93 CR598370 -2.87 SYNE2 -2.83 NDE1 -2.98 SGCE -2.92 A_24_P195164 -2.87 BCCIP -2.83 S100PBP -2.98 ATAD3B -2.92 TMUB1 -2.87 DLG1 -2.83 FLJ14346 -2.98 tcag7.1017 -2.92 HIPK3 -2.87 XPO6 -2.83 FJX1 -2.98 ZNF263 -2.92 ANKRD40 -2.87 PCBP2 -2.83 CCBL1 -2.98 ENST00000258451 -2.92 TCF25 -2.87 FLJ35220 -2.83 KHK -2.98 TCEA1 -2.92 ENST00000360796 -2.87 SR-A1 -2.82 C6orf167 -2.97 A_24_P118813 -2.92 RPRC1 -2.87 METTL9 -2.82 ABCB8 -2.97 AF229166 -2.92 USF2 -2.87 AK091836 -2.82 EZH2 -2.97 MLLT7 -2.92 MKNK2 -2.87 AF086335 -2.82 THC2268216 -2.97 ITGA5 -2.92 RP3-402G11.5 -2.86 AK057443 -2.82 WBP2 -2.97 FLJ40330 -2.91 NPR2 -2.86 RPL23 -2.82 LUC7L2 -2.97 DKC1 -2.91 SAAL1 -2.86 BC013250 -2.82 FUZ -2.97 ZNF530 -2.91 TMEM48 -2.86 CR591776 -2.82 SUV39H1 -2.97 AV749257 -2.91 A_23_P96017 -2.86 CXorf15 -2.82 BNIP3L -2.97 FOXP1 -2.91 GPT2 -2.86 76P -2.82 NONO -2.96 ZNF714 -2.91 LOC643159 -2.86 AFMID -2.82 XRCC3 -2.96 A_23_P255637 -2.91 FBXO11 -2.86 CNOT4 -2.82 AL117599 -2.96 ST7 -2.91 ENST00000320216 -2.86 CDC2L5 -2.82 SKIV2L2 -2.96 PPP2R2A -2.91 CAMTA2 -2.86 NUP210 -2.82 KRT18 -2.96 THC2315472 -2.91 PTPLAD1 -2.86 SLFN13 -2.82 DOCK7 -2.96 SPHK1 -2.91 ENOSF1 -2.86 LMAN1 -2.82 TBRG1 -2.96 C10orf46 -2.91 LOC442075 -2.86 SLC44A2 -2.82 GORASP2 -2.96 C6orf32 -2.90 EIF4A2 -2.86 APEX1 -2.82 FAM76B -2.96 SLC25A23 -2.90 SCNN1B -2.86 ITPR3 -2.82 NAGPA -2.96 TRIM26 -2.90 CD300A -2.86 KIAA0922 -2.82 A_24_P92411 -2.96 ZNF700 -2.90 UNC84A -2.86 LOC286161 -2.81 CNOT1 -2.96 JMJD2C -2.90 RFP2 -2.86 HUWE1 -2.81 GNAI2 -2.95 MUTED -2.90 CHST10 -2.85 HIAT1 -2.81 BF210146 -2.95 LOC150223 -2.90 REPS1 -2.85 ATG9A -2.81 VAT1 -2.95 LOC643454 -2.90 FXR1 -2.85 PPP2R5A -2.81 WBSCR1 -2.95 BCAT1 -2.90 SH2D5 -2.85 ENST00000372594 -2.81 SLC43A3 -2.95 GSTM1 -2.90 PSAP -2.85 BI050742 -2.81 TRIB3 -2.95 ACP2 -2.89 ZNF318 -2.85 DPP9 -2.81 LOC196394 -2.95 FSCN1 -2.89 CLK1 -2.85 NRF1 -2.81 BC041401 -2.94 LAT -2.89 KIAA1542 -2.85 PLD1 -2.81 M-RIP -2.94 TAGAP -2.89 SNTB1 -2.85 C22orf9 -2.80 CRSP7 -2.94 THC2280075 -2.89 HNRPA0 -2.85 ZFP36 -2.80 RASSF7 -2.94 MTMR12 -2.89 FH -2.85 ZNF267 -2.80

124

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2728) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) CRK -2.80 PTPN23 -2.77 PPP2R5B -2.73 TMEM87B -2.69 DAG1 -2.80 A_24_P213256 -2.77 ABL1 -2.73 NUP98 -2.69 H1FX -2.80 RAP1GDS1 -2.76 PPM1G -2.73 TNFRSF13B -2.69 NFRKB -2.80 A_24_P195621 -2.76 ABTB1 -2.73 C1orf80 -2.69 C1orf128 -2.80 HNRPU -2.76 BC035091 -2.73 ST13 -2.69 NME4 -2.80 ACADS -2.76 CTCF -2.72 A_24_P289504 -2.69 ZNF34 -2.80 CRKL -2.76 OSBPL11 -2.72 CENTB1 -2.69 AW029229 -2.80 USP21 -2.76 ENST00000381498 -2.72 GALNACT-2 -2.69 HMG20B -2.80 WDR46 -2.76 MGC16384 -2.72 BSPRY -2.69 ENST00000228360 -2.80 KIAA0738 -2.76 KIAA1919 -2.72 PIK3R5 -2.69 RANBP9 -2.80 PISD -2.76 AK123461 -2.72 USP10 -2.69 TCP1 -2.80 C17orf50 -2.76 AFG3L1 -2.72 THC2378504 -2.68 ENST00000323501 -2.80 ACSS1 -2.76 DULLARD -2.72 THC2366678 -2.68 LYPLA2 -2.80 KIAA1147 -2.76 ITGAL -2.72 THC2304728 -2.68 SNX17 -2.80 ESR2 -2.76 NPM1 -2.72 PIK3CB -2.68 THC2290002 -2.80 KIAA1794 -2.75 AP3S2 -2.72 ANKRD25 -2.68 NID1 -2.80 ACSL3 -2.75 AK130705 -2.72 SMAP1L -2.68 ABCC1 -2.80 RAB11FIP1 -2.75 DENND4A -2.72 CNDP2 -2.68 C6orf68 -2.80 09/01/2007 -2.75 ENST00000340855 -2.72 KIAA0515 -2.68 A_24_P375870 -2.80 GTPBP1 -2.75 ANAPC1 -2.72 ARHGAP27 -2.68 C14orf130 -2.80 ENST00000342584 -2.75 MAML1 -2.72 MAD2L2 -2.68 WNT10A -2.80 UBE2M -2.75 CHEK1 -2.71 SH2B3 -2.68 GBA2 -2.80 CPNE8 -2.75 BRD9 -2.71 C6orf72 -2.68 ST3GAL3 -2.80 ARCN1 -2.75 C2orf15 -2.71 ENST00000377538 -2.68 PRUNE -2.79 LOC399491 -2.75 C9orf93 -2.71 CGGBP1 -2.67 PALM2-AKAP2 -2.79 RBM13 -2.75 PARVB -2.71 LDLR -2.67 TOE1 -2.79 ADCK4 -2.74 MAT2A -2.71 GPSM2 -2.67 A_24_P455100 -2.79 WDR42A -2.74 SMPD4 -2.71 FMNL2 -2.67 AFF1 -2.79 A_24_P143653 -2.74 MTERFD1 -2.71 ADD1 -2.67 NFE2L3 -2.79 C21orf66 -2.74 A_32_P46700 -2.71 BC047032 -2.67 ATM -2.79 ENST00000319373 -2.74 SHQ1 -2.71 TNRC5 -2.67 A_24_P811954 -2.79 HPCAL1 -2.74 RIC8B -2.71 RASA4 -2.67 ALKBH5 -2.79 SLC43A2 -2.74 HNRPC -2.71 C6orf134 -2.67 AOF1 -2.79 PDCD6IP -2.74 PFKFB4 -2.71 BC009228 -2.67 ENST00000320831 -2.79 A_32_P224926 -2.74 DDX24 -2.71 SLC4A7 -2.67 THC2309960 -2.79 C3orf39 -2.74 ZNF598 -2.70 TM2D2 -2.67 BU153693 -2.79 C9orf91 -2.74 LOC90355 -2.70 SMARCA4 -2.67 A_24_P324250 -2.79 ENST00000308862 -2.74 SENP1 -2.70 USP25 -2.67 A_24_P349580 -2.78 LOC389833 -2.74 SUPT16H -2.70 SSX2IP -2.67 SLC38A2 -2.78 ARHGEF2 -2.74 A_24_P67268 -2.70 CYB5R3 -2.67 ACOT4 -2.78 MKLN1 -2.74 METT5D1 -2.70 TLK1 -2.67 UBTF -2.78 STAT5B -2.73 FADD -2.70 PPP1R9B -2.67 PIGS -2.78 KCNAB2 -2.73 AK129879 -2.70 RAC3 -2.67 DOCK6 -2.78 A_32_P72477 -2.73 LIMK2 -2.70 SLC3A2 -2.67 APLN -2.78 BC104430 -2.73 FCER2 -2.70 XRCC2 -2.67 BAZ2A -2.78 CARS -2.73 B4GALT2 -2.70 RARA -2.66 PPP1CC -2.78 A_24_P401392 -2.73 EIF4G1 -2.70 C10orf75 -2.66 FKBPL -2.78 RBPSUH -2.73 LOC401152 -2.70 ENST00000288548 -2.66 SBDS -2.78 PTPN4 -2.73 XPO7 -2.70 C8orf55 -2.66 TSR2 -2.77 MAP4 -2.73 SERBP1 -2.70 CTNNAL1 -2.66 ATP1A4 -2.77 FBXO32 -2.73 QSER1 -2.69 MYH9 -2.66 RIPK2 -2.77 RNF144 -2.73 A_24_P585660 -2.69 NARG1 -2.66 HDGF -2.77 PRPF4B -2.73 FAM80A -2.69 ZNF136 -2.66 THC2311196 -2.77 VGLL4 -2.73 DAGLBETA -2.69 TNIP1 -2.66 SPRED1 -2.77 CYP51A1 -2.73 CBLB -2.69 MTDH -2.66 CKAP5 -2.77 MGC22014 -2.73 CBX8 -2.69 LOC285074 -2.66 MAPK3 -2.77 DDX18 -2.73 BX648930 -2.69 DSCR3 -2.66 RSAD1 -2.77 EPS8L1 -2.73 BC033986 -2.69 BZW1 -2.66 CDC91L1 -2.77 A_24_P834646 -2.73 OMP -2.69 SLC16A9 -2.66

125

TABLE S-T6: BRCA1 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2728) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) THC2403568 -2.66 THC2312955 -2.61 GMCL1 -2.57 STX18 -2.53 TPM3 -2.66 MPDU1 -2.61 TYRO3 -2.57 SH3GL1 -2.53 ZC3H14 -2.66 PTPN7 -2.61 COCH -2.57 CD47 -2.53 A_24_P780609 -2.65 KIAA0406 -2.61 OSBP -2.57 SPRY4 -2.53 BC036622 -2.65 NTNG2 -2.61 ZNF57 -2.57 LOC440353 -2.53 GNB5 -2.65 ENST00000321792 -2.61 EFCAB4B -2.57 NFKBIL1 -2.53 CTDSPL2 -2.65 RRN3 -2.61 SOCS4 -2.57 OFD1 -2.53 AGPAT6 -2.65 RBBP8 -2.61 MIDN -2.57 LOC338756 -2.52 HIPK1 -2.65 ATG16L1 -2.61 NUP93 -2.57 BAZ1B -2.52 RSL1D1 -2.65 KPNA4 -2.61 GRLF1 -2.57 CD151 -2.52 A_24_P187154 -2.65 VDAC3 -2.61 TNF -2.57 C6orf96 -2.52 WDR36 -2.65 NARF -2.61 PAX5 -2.57 A4GALT -2.52 MOAP1 -2.65 CHD4 -2.61 INTS4 -2.56 CRYBB2 -2.52 THC2267456 -2.64 C6orf64 -2.60 BCL2L13 -2.56 ENST00000376356 -2.52 LASS2 -2.64 D80006 -2.60 MLLT10 -2.56 NELF -2.52 CAPZB -2.64 SERGEF -2.60 SCYL1 -2.56 TOMM34 -2.52 MYST1 -2.64 ANKRD32 -2.60 MAPK12 -2.56 PAPD4 -2.52 ARAF -2.64 MPP1 -2.60 ATP2A3 -2.56 OGDH -2.52 PARP10 -2.64 BRCA1 -2.60 ENST00000374537 -2.56 ZCCHC7 -2.52 CXXC1 -2.64 POGK -2.60 GCET2 -2.56 USP7 -2.52 PELP1 -2.64 FCMD -2.60 TTF2 -2.56 CR610131 -2.52 CLCN2 -2.64 PEX5 -2.60 PCQAP -2.56 SLC4A2 -2.52 ZNF84 -2.64 A_24_P401150 -2.60 ENST00000374717 -2.56 NFATC3 -2.52 SCLY -2.64 A_24_P161733 -2.60 PPP2R1A -2.56 KIAA0831 -2.52 FLJ38984 -2.64 NIPA1 -2.60 TRIT1 -2.55 THC2364375 -2.52 FADS1 -2.64 KLC4 -2.60 DDX3Y -2.55 A_24_P229766 -2.52 TADA1L -2.64 ATP8B2 -2.60 LOC63920 -2.55 A_32_P8806 -2.52 FLJ11171 -2.64 C4orf9 -2.60 FAM119A -2.55 P15RS -2.52 LRAP -2.64 VPRBP -2.60 SNX16 -2.55 A_24_P306614 -2.52 NRBP1 -2.64 C10orf6 -2.60 ZNFN1A2 -2.55 SSTR2 -2.52 GFPT1 -2.63 SNRPD1 -2.60 KBTBD2 -2.55 AK026225 -2.52 PFDN2 -2.63 THC2375588 -2.59 GEM -2.55 BC087732 -2.51 GSTM4 -2.63 KIAA0649 -2.59 CNP -2.55 MAX -2.51 TLR10 -2.63 USP48 -2.59 FAM113A -2.55 PCOLN3 -2.51 ENST00000379841 -2.63 A_24_P187174 -2.59 SNRP70 -2.55 C1orf106 -2.51 NHP2L1 -2.63 RIN3 -2.59 PARP14 -2.54 SNIP1 -2.51 ATXN2L -2.63 MLXIP -2.59 ZDHHC6 -2.54 TMCC1 -2.51 THUMPD3 -2.63 BF895757 -2.59 SETD5 -2.54 LOC389831 -2.51 LOC389517 -2.63 SLC16A13 -2.59 A_24_P349869 -2.54 HCP1 -2.51 C13orf25 -2.63 TRAFD1 -2.59 ACLY -2.54 AK056809 -2.51 HES6 -2.63 HSPA5 -2.59 JMJD1A -2.54 IFT52 -2.51 DDEF1 -2.63 A_24_P418786 -2.59 UBE3A -2.54 COG3 -2.51 MUTYH -2.63 USP5 -2.58 RAD23A -2.54 IQCC -2.51 AK026675 -2.63 PEO1 -2.58 MGC4268 -2.54 C15orf39 -2.51 ARHGEF6 -2.63 GLI4 -2.58 TMEM18 -2.54 L40520 -2.51 AK022020 -2.63 THC2373821 -2.58 AK131288 -2.54 NR1H2 -2.51 RHPN2 -2.63 VPS13B -2.58 C9orf64 -2.54 HMGCS1 -2.51 RGS14 -2.63 CDC7 -2.58 CR617352 -2.54 AK022252 -2.51 NUDC -2.62 A_24_P534290 -2.58 CDKN3 -2.54 MAPRE2 -2.51 ENST00000367942 -2.62 PPRC1 -2.58 OPRS1 -2.54 PYGB -2.51 EIF4B -2.62 SNRPA -2.58 FLJ11021 -2.53 PRPF38B -2.51 BQ072652 -2.62 TNRC6B -2.58 STK38 -2.53 SLC25A37 -2.51 FLJ31818 -2.62 MPP6 -2.58 A_24_P126731 -2.53 YPEL1 -2.50 NIPSNAP1 -2.62 A_23_P99731 -2.58 IGHMBP2 -2.53 CR606271 -2.50 SLC35A3 -2.62 AK026811 -2.58 DPP8 -2.53 DENND4C -2.50 TMEM38A -2.62 MS4A1 -2.58 03/01/2007 -2.53 RHOT1 -2.50 TBCD -2.62 PHF19 -2.58 E2F4 -2.53 A_23_P96035 -2.50 RAD50 -2.62 TM6SF1 -2.58 GTPBP2 -2.53 ZNF519 -2.50

126

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE UP-REGULATED GENES (2012) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) HES2 9.85 FLJ11259 5.17 C16orf5 4.27 THC2314369 3.83 ASTN2 9.67 ACTA2 5.17 C1orf57 4.26 SIDT1 3.83 SULF2 9.49 LOC133874 5.17 KIAA1219 4.25 LOC257396 3.81 A_32_P52153 8.85 THC2340838 5.15 FOLR2 4.24 NIN 3.80 CN430223 7.98 SDC1 5.11 CAND1 4.24 THC2376737 3.79 SPATA18 7.98 SLC4A11 5.08 STK4 4.22 A_23_P17152 3.79 AK092083 7.94 PHLDA3 5.08 FOSL1 4.22 ZBTB20 3.78 THC2429167 7.59 FLJ13576 5.04 GCH1 4.20 AP1G1 3.78 THC2437069 7.57 PRKY 5.03 ENST00000361453 4.19 ZMAT3 3.78 CABYR 7.53 A_23_P64962 5.01 A_24_P281853 4.18 MFSD4 3.77 SESN1 7.44 BRMS1L 4.96 BF960555 4.18 ENST00000235345 3.75 ENST00000377836 7.39 KCTD1 4.94 SPARC 4.18 TNFSF10 3.73 MDM2 7.35 CASK 4.86 PLXNB2 4.16 PDGFRA 3.72 TRIM22 7.35 A_32_P225768 4.84 SFXN5 4.14 CD242823 3.72 AK024898 7.30 TNFSF4 4.82 ENST00000368491 4.12 Ells1 3.72 PLK2 7.21 MAMDC4 4.79 CXorf43 4.12 ALOX5 3.71 RHO 7.16 ZNF79 4.78 CB250445 4.12 THC2310998 3.71 VWCE 7.09 BC053363 4.77 LOC339768 4.11 NCSTN 3.70 PAPLN 6.81 FEZ1 4.76 ENST00000378179 4.11 ENST00000329309 3.68 C8orf38 6.79 LOC653374 4.72 ENST00000361227 4.10 FAS 3.68 FBXO22 6.75 MOSPD1 4.69 MGC70863 4.10 LOC286254 3.68 LOC134147 6.72 THC2404359 4.69 STARD4 4.08 MYO5A 3.68 TNFRSF10C 6.69 TP53INP1 4.69 AK093416 4.07 DHTKD1 3.66 AI500335 6.57 C20orf161 4.68 DB518505 4.06 PSTPIP2 3.66 ANKRD47 6.52 PRKX 4.67 DLGAP4 4.06 THC2322041 3.66 LOC340109 6.46 THC2450504 4.67 XPC 4.05 CR603195 3.65 SLC6A19 6.38 BC030100 4.63 BTBD14A 4.04 ABCA12 3.64 BC040303 6.35 ANKRA2 4.63 ITPR2 4.04 C19orf6 3.64 GLS2 6.34 PTP4A1 4.63 ENST00000369572 4.03 CSNK1D 3.64 MGC5370 6.32 MYO1A 4.58 UBR1 4.02 SLC30A3 3.63 SLC35E3 6.30 PHACTR4 4.58 TNFRSF10B 4.01 NSF 3.63 GIPR 6.28 THC2315966 4.58 DQX1 4.01 AY007156 3.63 WDR63 6.26 ARHGEF3 4.58 CFD 4.00 PHF16 3.63 LIF 6.25 SLC7A6 4.54 PPM1D 4.00 MAP3K7IP3 3.62 TRIAP1 6.23 THC2273762 4.54 A_24_P531074 3.99 OR4D2 3.62 SMPD3 6.22 PODXL 4.52 THC2406514 3.99 GPR172B 3.61 THC2343350 6.11 SART2 4.51 SPR 3.98 HNRPH2 3.60 AF144054 6.09 GRHL3 4.48 LENG8 3.98 OASL 3.60 SARDH 6.02 PRSS36 4.47 HEMK1 3.93 BC066984 3.60 TM7SF3 6.00 GLT8D3 4.47 THC2266610 3.93 BLCAP 3.60 IL10RB 5.98 THC2316748 4.47 A_23_P206568 3.93 FKSG2 3.60 ZNF385 5.90 ENST00000340381 4.45 ARHGEF15 3.93 THC2248354 3.59 C1orf183 5.90 BLOC1S2 4.45 FAM83H 3.93 PCNXL2 3.59 POLH 5.71 ITM2A 4.44 HTR7P 3.92 USP33 3.59 TP53I3 5.71 AK056245 4.43 DCP1B 3.91 C11orf41 3.58 AA887631 5.69 A_23_P6514 4.42 A_24_P127042 3.91 SLC35D1 3.58 MDS025 5.68 A_32_P12282 4.41 THC2277837 3.91 CHST6 3.56 PLAT 5.61 RRM2B 4.38 PHTF1 3.90 LOC158863 3.55 NTN1 5.60 AK026194 4.35 BU567832 3.90 TRIP6 3.55 HGS 5.57 THC2439499 4.35 DEPDC5 3.89 DIRC1 3.55 CEACAM1 5.57 ITPKC 4.31 FAM41C 3.89 PGPEP1 3.54 ASCC3 5.45 ADAL 4.31 DCUN1D3 3.89 THC2374442 3.54 AK026338 5.44 ITGAM 4.30 A_32_P31206 3.88 THC2281539 3.54 ANK1 5.41 ENST00000382004 4.30 ZFP41 3.87 C1orf26 3.54 SMTN 5.39 NAP1L5 4.29 RGS12 3.87 ENST00000298453 3.54 C3orf23 5.36 OR11A1 4.29 A_24_P916853 3.87 HSDL2 3.54 FHL2 5.35 BRUNOL6 4.28 ENST00000377093 3.86 NT5DC1 3.52 C12orf5 5.32 ST5 4.28 RRAD 3.86 ZC3HAV1 3.52 LRDD 5.29 METTL7A 4.28 DUSP18 3.85 MAGEL2 3.52 SYTL1 5.22 TP53AP1 4.28 BC035180 3.84 TAAR5 3.52

127

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE UP-REGULATED GENES (2012) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) BCAS2 3.51 PHF6 3.32 SLFN5 3.15 KIAA1370 3.01 MECP2 3.51 AK026368 3.31 A_24_P75994 3.15 IGSF9 3.01 TNFRSF10A 3.51 ROD1 3.31 BC035751 3.15 CYP3A5 3.00 SLC22A5 3.51 RBL2 3.31 ATP6V1A 3.14 THC2317830 3.00 ZFP90 3.49 LOC653483 3.30 CSPG2 3.14 BC041926 3.00 AK056119 3.49 THC2238625 3.30 CROT 3.13 KIAA1881 3.00 CYB5R1 3.49 DCUN1D1 3.30 KIAA0999 3.13 TANC1 3.00 GM2A 3.48 IGBP1 3.29 CR609588 3.13 HRAS 3.00 AK097322 3.48 COQ10A 3.29 DZIP3 3.13 MAX 3.00 BTBD10 3.47 APBB3 3.29 CDK5R2 3.13 PSKH1 2.99 KIAA0888 3.47 THC2280109 3.29 LOC645431 3.12 CRIPAK 2.99 SESN2 3.46 TOB1 3.29 LOC51136 3.12 UBE2D1 2.99 EFNB1 3.45 PFTK1 3.28 ENST00000367545 3.12 SLC26A11 2.99 ISG20L1 3.45 ENC1 3.28 IL13RA1 3.12 CASQ1 2.99 WDR19 3.45 EMX1 3.28 C17orf85 3.11 ATG4A 2.99 C18orf56 3.45 CCNK 3.28 A_32_P108748 3.11 CCDC113 2.99 FLJ10781 3.44 TTC21B 3.27 RABGGTA 3.11 SLC2A11 2.99 ENST00000372072 3.44 MGC70870 3.27 A_24_P144054 3.11 PARD6G 2.98 CP110 3.44 ENST00000311275 3.26 CARD12 3.11 LCE1F 2.98 AK092875 3.43 FUCA1 3.25 ZNF435 3.10 THC2310680 2.98 C20orf12 3.43 TSGA10 3.25 PHF20L1 3.10 SERPINB8 2.98 SGPL1 3.43 MRPL27 3.25 OAT 3.10 SNX22 2.98 LOC57149 3.42 PPFIBP2 3.24 BF718543 3.10 FCER1G 2.98 ENST00000290607 3.42 MAWBP 3.24 RASGRF1 3.10 FLJ40176 2.98 LONPL 3.42 PIGH 3.24 AKAP7 3.10 ZNF597 2.98 SLC22A15 3.42 NDFIP2 3.24 CREB3L1 3.10 PSMD10 2.98 MYO6 3.42 NCOA3 3.24 BBS1 3.10 MIB2 2.97 RRAGA 3.41 THC2376817 3.23 SOLH 3.09 CLEC4M 2.96 ZNF337 3.41 B3GNT8 3.23 PTMS 3.09 A_23_P207049 2.96 SLC13A2 3.41 TCP11L1 3.23 EI24 3.09 PROCR 2.96 ENST00000380357 3.41 POLR2K 3.23 LOC645287 3.09 LOC201229 2.96 OR5L2 3.40 SLCO2B1 3.23 ENST00000332804 3.08 KIAA0284 2.96 RBM16 3.40 PRKAB2 3.22 C3orf50 3.08 PCGF3 2.96 CRYZL1 3.40 AV739766 3.22 ENST00000374860 3.08 FLJ32065 2.96 AK055641 3.40 PTPN1 3.22 SPTY2D1 3.08 FAM21C 2.96 CCDC92 3.39 KIAA0329 3.22 SCRN3 3.08 C4orf24 2.96 PRKAB1 3.39 THC2436415 3.22 ENST00000308118 3.07 A_24_P635355 2.95 LOC92017 3.39 ANKRD46 3.22 A_24_P144487 3.07 PTPLA 2.95 SUCLA2 3.39 ZNF425 3.22 AK125129 3.07 TRIP4 2.95 GADD45A 3.39 ITM2B 3.21 UBQLN1 3.07 C14orf28 2.95 COL4A1 3.39 ADAM10 3.21 CCAR1 3.06 C1orf25 2.95 C2orf13 3.39 AK092942 3.21 RAPH1 3.05 CR626222 2.95 ZNF219 3.39 MSI2 3.21 THC2427841 3.05 LMBR1 2.95 PLEKHQ1 3.38 C1orf42 3.20 S100A11 3.05 FCGBP 2.95 THC2407694 3.38 MAP4K4 3.20 APOBEC3G 3.05 EID-3 2.94 UHMK1 3.37 KCNN3 3.19 ZMAT2 3.04 UBE4B 2.94 ENST00000367233 3.37 KIAA1005 3.19 BX093417 3.04 PKD2 2.94 CHURC1 3.37 FLJ10815 3.19 TMEM30A 3.04 RFX5 2.93 FAM98C 3.36 CSNK1G1 3.19 NISCH 3.04 ZNF253 2.93 AL049387 3.36 FRK 3.18 KIDINS220 3.04 BC035518 2.93 TMEM150 3.36 C19orf46 3.18 CDKN1A 3.04 SLFN12 2.93 A_24_P127462 3.35 AK098185 3.18 A_32_P40375 3.04 THC2257370 2.93 AF086139 3.35 MICB 3.17 ENST00000360523 3.04 C12orf49 2.92 FAM46A 3.35 BX103476 3.17 ENST00000369739 3.03 FLJ20294 2.92 SNX13 3.34 LPXN 3.16 NIPSNAP3A 3.03 LOC644053 2.92 ADRB2 3.34 THC2408506 3.16 A_23_P216071 3.03 KLHL7 2.92 TRIM35 3.34 ZNF30 3.16 HHAT 3.02 NDUFV3 2.92 UGT2B10 3.33 SLC22A18AS 3.16 AL832758 3.02 BE672985 2.91 GAMT 3.33 ANUBL1 3.15 APBA3 3.02 SCNN1D 2.91 BACE1 3.32 RETSAT 3.15 A_24_P850187 3.01 KLHL12 2.91

128

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE UP-REGULATED GENES (2012) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) SRA1 2.90 BMP2K 2.81 FGL2 2.72 C6orf173 2.64 CHST2 2.90 FBXO28 2.81 THC2289056 2.72 ENST00000358916 2.64 X75962 2.90 C9orf6 2.81 THC2343435 2.72 FRG1 2.64 PHF15 2.90 OR10H2 2.81 FASTKD5 2.72 THC2401540 2.64 PCGF5 2.90 CR627133 2.80 FGF3 2.72 LOC399744 2.64 AK096685 2.90 A_32_P23872 2.80 PDE4C 2.72 ENST00000377548 2.64 CABC1 2.89 PIGR 2.80 LOC400509 2.72 HEAB 2.64 SLC34A1 2.89 KIAA1908 2.80 AL137705 2.71 MXD4 2.64 RPA4 2.89 SAMD9 2.80 A_24_P827794 2.71 GP9 2.63 PTPRE 2.89 CCDC98 2.80 MT3 2.71 THC2437618 2.63 CCDC32 2.89 CECR1 2.80 GSK3B 2.71 AMZ2 2.63 BX101252 2.89 MFSD1 2.80 THC2366161 2.71 KIAA1109 2.63 REEP2 2.88 CD274 2.79 ENST00000301807 2.71 BTG2 2.63 FZD6 2.88 XPR1 2.79 CA13 2.71 WDR60 2.63 THC2275950 2.88 SAC3D1 2.79 GPR109B 2.71 ENST00000373335 2.63 ID3 2.88 TncRNA 2.78 CEP135 2.70 CR619760 2.62 ARL6IP5 2.88 PCNX 2.78 NPAL3 2.70 ABTB2 2.62 MARVELD3 2.88 DKFZp667M2411 2.78 CENTB2 2.70 MUC20 2.62 BAT2D1 2.88 RALA 2.78 MRPL49 2.70 OSBPL3 2.62 STX6 2.88 TNFRSF1A 2.78 PLK3 2.69 DNAJA5 2.62 THC2433060 2.87 LOC147650 2.77 THC2277201 2.69 DEPDC6 2.62 BI029121 2.87 TMEM19 2.77 ENST00000238789 2.69 KIFAP3 2.62 ENST00000367590 2.87 TRAPPC2 2.77 AF085351 2.69 PML 2.62 LOC151534 2.87 FLJ26850 2.77 TRFP 2.69 ACOT2 2.61 MAP3K12 2.87 MAD1L1 2.77 CBWD2 2.69 AK090499 2.61 AY203961 2.87 DMXL1 2.77 AADACL1 2.68 ZNF690 2.61 GRIN2C 2.86 SGCB 2.77 ZDHHC7 2.68 MMP11 2.61 THC2279364 2.86 PRIMA1 2.77 BE138567 2.68 HCFC2 2.61 CD511705 2.86 EAF1 2.77 ULK1 2.68 CCDC55 2.61 AK123333 2.86 MMACHC 2.76 SMAD1 2.68 OR2H1 2.61 TMEM129 2.86 OR7E47P 2.76 SGOL1 2.68 EIF2AK4 2.60 FLJ30092 2.86 MYBPC2 2.76 THC2324430 2.68 CEP164 2.60 PLA2G4D 2.85 LOC51252 2.76 CCL24 2.68 THC2281731 2.60 KRT17 2.85 GRAMD3 2.76 TMEM117 2.67 POMT1 2.60 AHRR 2.85 AL831999 2.76 AW977527 2.67 LOC253039 2.60 C11orf24 2.85 FRMPD2 2.76 THC2439806 2.67 SUMF1 2.60 ENST00000312289 2.85 GYG1 2.76 FLJ13231 2.67 A_32_P151782 2.60 KLHL3 2.85 ENST00000317868 2.76 ITGAV 2.67 ZMYM1 2.59 C5orf5 2.85 EEA1 2.75 FAM43B 2.67 LOC647090 2.59 MLLT1 2.85 LOC340061 2.75 LOC158960 2.67 MGC34796 2.59 ACOX1 2.84 3.8-1 2.75 THC2286151 2.66 ADH5 2.59 SLTM 2.84 ITFG1 2.75 DISP1 2.66 FCRLM1 2.59 ALCAM 2.84 OR8B8 2.75 GOLGB1 2.66 ZNFN1A4 2.59 NOTCH1 2.84 TSG101 2.75 TEAD3 2.66 LCE2A 2.59 ADRBK2 2.84 DIRC2 2.75 HERC4 2.66 FLJ39370 2.59 AMPD3 2.84 DUSP13 2.75 TMEM138 2.66 GGCX 2.58 OR7E91P 2.83 HERC5 2.75 SGTB 2.65 ZZEF1 2.58 KRTAP10-10 2.83 C14orf159 2.75 AK097080 2.65 BX107298 2.58 ZNF658 2.83 LYST 2.74 CR606637 2.65 HARS 2.58 ZCWPW1 2.83 MRRF 2.74 FLJ23569 2.65 ZNF407 2.58 LSM6 2.83 TBC1D5 2.74 OXSM 2.65 GFAP 2.57 CXX1 2.83 LAX1 2.74 ARMCX6 2.65 CRHR1 2.57 SIDT2 2.82 TMEM68 2.74 A_24_P229728 2.65 A_32_P131870 2.57 THEM5 2.82 FLJ36868 2.74 ALG8 2.65 A_24_P341078 2.57 RAB8B 2.82 LOC646626 2.74 A_32_P187143 2.65 THC2261399 2.57 OBFC2A 2.82 MTMR3 2.74 BM989484 2.65 MORN2 2.57 TRPM5 2.82 GGTL4 2.74 BBS4 2.64 HELB 2.57 ATF3 2.82 KDELC1 2.73 A_32_P5628 2.64 STK17A 2.57 THC2336549 2.82 BX648950 2.73 LHX4 2.64 C12orf45 2.57 GUSBL2 2.81 BAZ2B 2.72 CTGLF1 2.64 LOC441245 2.56

129

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE UP-REGULATED GENES (2012) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) ATXN3 2.56 ATP6V1H 2.51 COL23A1 2.44 A_32_P177097 2.38 RP5-1022P6.2 2.56 PEX11B 2.51 BNIP2 2.44 GRRP1 2.38 ZNF655 2.56 PLEKHF1 2.51 BF446608 2.44 PGBD2 2.38 KIAA0323 2.56 ENST00000375672 2.50 HGF 2.44 TRIM32 2.37 DHFRL1 2.56 SCYL3 2.50 TNS3 2.43 A_23_P129405 2.37 THC2308747 2.56 ENST00000277575 2.50 FLJ30064 2.43 CACNA1C 2.37 DKFZP434A0131 2.56 LOC283551 2.50 PTEN 2.43 BRWD2 2.37 TMEM118 2.56 WDR59 2.50 AK023557 2.43 BCORL1 2.37 MKNK1 2.56 NOTCH2 2.50 LOC442421 2.43 KIAA0513 2.37 BE537483 2.56 ZFR 2.50 DKFZp547C195 2.43 RSHL2 2.37 DR1 2.56 TMEM142C 2.50 ZNF135 2.43 SAMD9L 2.37 OR5H1 2.56 RUNDC2A 2.50 PARN 2.43 C20orf112 2.37 PCDHGA8 2.55 OR7E24 2.49 AK055915 2.42 HOXB6 2.37 E2F7 2.55 MAP2K6 2.49 A_23_P329062 2.42 NOD3 2.37 TRAM2 2.55 POLK 2.49 RP11-217H1.1 2.42 DAMS 2.37 BC038355 2.55 ZC3H12A 2.49 ENST00000378953 2.42 BC037328 2.37 CCNG1 2.55 MIA3 2.49 ENST00000378536 2.42 TMEM32 2.36 REST 2.55 FPGT 2.48 ENST00000324982 2.42 C1orf88 2.36 TOP2A 2.55 ZNF469 2.48 ATP6AP1 2.42 ATN1 2.36 BC037740 2.55 PNMA1 2.48 C6orf35 2.42 SCGB1A1 2.36 ORAOV1 2.55 ENST00000336283 2.48 SYK 2.42 SERTAD1 2.36 RBM41 2.54 OR7E13P 2.48 DMD 2.42 C1orf186 2.36 ZNF195 2.54 IL27 2.48 THC2342491 2.42 ATG5 2.36 PIN4 2.54 PPARD 2.48 PHF8 2.41 BC031316 2.35 DKFZP564B147 2.54 ARID5B 2.48 BIRC4BP 2.41 CTNND1 2.35 NDUFA5 2.54 SYNE1 2.48 UTP14C 2.41 MT1A 2.35 C20orf108 2.54 TBC1D24 2.48 VTI1A 2.41 ZNF33B 2.35 CDC42EP4 2.54 LRRC25 2.48 LOC401357 2.41 APOL2 2.35 PALM 2.53 WFS1 2.48 TMEM128 2.41 ENST00000340158 2.35 GALNT10 2.53 ATG4C 2.48 ENST00000186436 2.41 SLC9A3R1 2.35 DDX58 2.53 ID2 2.48 PDE4B 2.40 C21orf91 2.35 ENST00000261569 2.53 AQP10 2.48 TMEM15 2.40 A_32_P101653 2.34 BU685299 2.53 HIF1A 2.47 DPH5 2.40 ENST00000328644 2.34 RNU12 2.53 RICS 2.47 THC2395899 2.40 ENST00000278949 2.34 ARHGAP6 2.53 ENST00000262525 2.47 ACAD8 2.40 ZFHX2 2.34 ENST00000317633 2.53 ENST00000299756 2.47 PRKCI 2.40 MRPL39 2.34 AK098422 2.52 MARVELD1 2.47 RAI1 2.40 STX1A 2.34 CES2 2.52 RFXDC2 2.47 MT1F 2.40 KIAA1468 2.34 TLE4 2.52 EPM2AIP1 2.47 A_32_P30187 2.40 KIAA1279 2.34 THC2314566 2.52 LOC653857 2.47 LOC134145 2.40 TBXA2R 2.34 ELF5 2.52 VPS45A 2.47 TM9SF1 2.39 PRKACB 2.34 LOC253981 2.52 CIAPIN1 2.46 STARD5 2.39 FBXO38 2.34 C21orf81 2.52 FLJ21657 2.46 CASC5 2.39 GNG4 2.34 C3orf9 2.52 CR603184 2.46 FLJ32255 2.39 NSD1 2.34 HELZ 2.52 P2RY10 2.46 DBT 2.39 SERPINF1 2.33 TMEM2 2.52 KIF1B 2.46 CLCN3 2.39 AF086329 2.33 CHD9 2.52 BE005242 2.45 KIAA0408 2.39 THC2335955 2.33 WFDC2 2.52 FAM49A 2.45 THC2411433 2.39 RAB43 2.33 SRP19 2.52 SEC61A2 2.45 FLJ10769 2.39 PARP14 2.33 MICAL-L2 2.52 LOC400642 2.45 TNRC6A 2.39 LAMC1 2.33 THC2440818 2.52 THC2314058 2.45 ANKRD20A2 2.39 LLGL2 2.33 BG108194 2.51 ENST00000374390 2.45 THC2440228 2.39 ENST00000361567 2.33 CASP9 2.51 RGL1 2.45 ABAT 2.38 TSPYL3 2.33 THC2287925 2.51 BAZ1A 2.45 TM7SF2 2.38 LOC150763 2.33 CRHR2 2.51 FOXO1A 2.45 C1orf124 2.38 TTC3 2.32 THC2364821 2.51 A_24_P670147 2.44 PEPD 2.38 ABHD3 2.32 AA878126 2.51 ENST00000330899 2.44 LOC344595 2.38 RHBDD1 2.32 NALP1 2.51 THC2390306 2.44 AK090897 2.38 ZZZ3 2.32 PPFIA1 2.51 LOC645919 2.44 LRRIQ2 2.38 ENST00000333731 2.32 IFI6 2.51 THC2373429 2.44 PTPRN2 2.38 INHBC 2.31

130

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE UP-REGULATED GENES (2012) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) TRAK1 2.31 AK057956 2.26 ENST00000382592 2.21 DUSP14 2.15 PTPN22 2.31 C9orf97 2.26 SPAST 2.21 DDX52 2.15 C10orf32 2.31 FAM49B 2.25 NOTCH2NL 2.21 CENTA2 2.15 C1orf102 2.31 CXCL5 2.25 MTHFR 2.21 VPS18 2.15 RABGAP1 2.31 CR627148 2.25 A_24_P170025 2.21 STXBP3 2.15 CYB5D2 2.31 ABHD13 2.25 ZCSL2 2.20 ATP10D 2.15 WWP1 2.31 N4BP2 2.25 TRIM38 2.20 PFDN4 2.15 THC2284657 2.31 C6orf170 2.25 FOXC1 2.20 AK090827 2.15 FNDC6 2.31 LOR 2.25 CR590163 2.20 IHPK2 2.15 TTC12 2.31 BC017972 2.25 THC2407434 2.20 CR619772 2.15 VSIG9 2.30 EXOC7 2.25 CR590071 2.20 TLR7 2.15 STOX2 2.30 MPEG1 2.24 RP4-742C19.3 2.20 PARC 2.15 SF3A1 2.30 BE646426 2.24 KRTAP5-8 2.20 PARP9 2.15 MACF1 2.30 ZNF286 2.24 LOC642299 2.20 MYO1E 2.15 DCTN6 2.30 WDR48 2.24 ENST00000380021 2.20 A_24_P346859 2.15 MGC16385 2.30 KIAA1429 2.24 CASP10 2.20 NAGS 2.15 CK300181 2.30 AK129838 2.24 AJ420487 2.19 BC029255 2.15 SMAD5 2.30 GNPTAB 2.24 AK123446 2.19 A_24_P204015 2.14 TTC18 2.30 FASTKD2 2.24 SUOX 2.19 C1orf164 2.14 SLC7A6OS 2.29 LOC644799 2.24 TAF9 2.19 SLC43A2 2.14 C14orf167 2.29 ZNF234 2.24 PRKAG2 2.19 WDR42A 2.14 RAGE 2.29 ACSS1 2.24 PRDM1 2.19 HERC6 2.14 ANKRD17 2.29 LOC440396 2.23 HSPC048 2.19 MGC10471 2.14 ZCCHC10 2.29 AK092791 2.23 TMEM85 2.19 FRY 2.14 LRP10 2.29 LOC643201 2.23 KIAA1443 2.19 USP15 2.14 DKFZP564C196 2.29 KIAA1632 2.23 SMEK2 2.19 ENST00000339867 2.14 LOC150759 2.29 CNR1 2.23 C6orf162 2.19 FLJ36031 2.14 AK022038 2.29 PDLIM4 2.23 A_32_P139021 2.19 MTM1 2.14 A_24_P882309 2.29 THC2444653 2.23 AK094629 2.19 YES1 2.14 ENST00000366751 2.29 BF576096 2.23 SRF 2.19 AQP2 2.14 MRPL10 2.28 GVIN1 2.23 CCPG1 2.19 ENST00000316634 2.14 COX7B 2.28 C20orf74 2.23 RBED1 2.18 ENST00000329367 2.14 KIAA1586 2.28 MYO9B 2.23 COL5A2 2.18 AF009267 2.13 THC2392085 2.28 NDFIP1 2.23 THC2377877 2.18 TSHB 2.13 THC2311946 2.28 GNPAT 2.23 C14orf108 2.18 BM561346 2.13 BC019824 2.28 IBRDC3 2.23 PNKD 2.18 A_24_P221105 2.13 CCDC120 2.28 PDPK1 2.23 AK024921 2.18 C1QDC1 2.13 THC2340803 2.28 A_24_P410256 2.23 CAPN10 2.18 ENST00000322831 2.13 GARNL1 2.28 MLKL 2.23 CD99L2 2.18 SLC25A20 2.13 STAT4 2.28 LRP6 2.23 ENST00000263956 2.18 IL1B 2.13 FLJ35934 2.28 ZDHHC16 2.22 AK026195 2.18 ATF6 2.13 THC2318533 2.27 OSRF 2.22 SLC39A13 2.17 BX115105 2.12 PMS2L1 2.27 GHSR 2.22 PTPN2 2.17 THEM4 2.12 POP1 2.27 SLC6A4 2.22 DNAJC16 2.17 LATS2 2.12 A_24_P187365 2.27 MGC23909 2.22 FMO4 2.17 MEG3 2.12 MPHOSPH1 2.27 FAM84A 2.22 PIGG 2.17 CB240572 2.12 DNHD1 2.27 SUV420H1 2.22 PTPLAD2 2.17 SERF1B 2.12 LYK5 2.27 KPNA5 2.22 BX111592 2.17 TMEM106B 2.12 CR617033 2.27 A_23_P28743 2.22 PRCP 2.17 TULP4 2.12 CFL2 2.27 ZNRF2 2.22 EIF2C1 2.17 C14orf103 2.12 A_23_P28397 2.27 KIAA0467 2.21 DAAM1 2.17 TTC23 2.12 CUTL1 2.27 C20orf44 2.21 CASC4 2.16 MR1 2.12 ARL6IP 2.27 FKSG44 2.21 DDX26B 2.16 C5orf21 2.12 BBS10 2.26 KIAA1212 2.21 PAFAH1B1 2.16 SLC39A11 2.12 CREB1 2.26 LOC440350 2.21 KIAA1826 2.16 AK057576 2.12 LOC128977 2.26 LRRK2 2.21 LRRC2 2.16 KIAA1383 2.12 EPB41L5 2.26 NACAP1 2.21 GCC2 2.16 TTYH3 2.11 ENST00000330044 2.26 AL079999 2.21 THC2442304 2.16 RNF121 2.11 BE968596 2.26 PDLIM5 2.21 FLJ10099 2.16 FAM117A 2.11 DHRS1 2.26 A_24_P196024 2.21 UBPH 2.16 ALAS1 2.11

131

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE UP-REGULATED GENES (2012) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) DST 2.11 RPS6KC1 2.07 KIAA1840 2.02 THC2312274 1.98 C1orf117 2.11 LTBP2 2.06 RANBP10 2.02 GGT2 1.98 CLASP1 2.11 PPP1R9B 2.06 C1orf182 2.02 CDK6 1.98 ENST00000316294 2.11 THC2377221 2.06 DOCK10 2.02 SLC17A7 1.98 PCNP 2.11 KIAA1737 2.06 THC2337124 2.02 PLCL2 1.98 MTRF1L 2.11 MED19 2.06 FMNL3 2.02 ENST00000223618 1.98 LOC646564 2.11 ENST00000324677 2.06 C19orf12 2.02 KIAA1012 1.98 ENST00000356572 2.10 CYLD 2.06 ABCD3 2.02 ATP11B 1.98 RAB38 2.10 A_24_P306788 2.06 UBQLN2 2.01 FLJ12688 1.98 THC2436814 2.10 NPC2 2.06 CR603982 2.01 XTP7 1.98 NRD1 2.10 NF1 2.06 DMPK 2.01 RP11-78J21.1 1.97 ZC3H14 2.10 THC2439773 2.06 A_24_P230009 2.01 AK098749 1.97 AI076466 2.10 ATG4B 2.05 U94903 2.01 SPEN 1.97 THC2280976 2.10 PPM1B 2.05 THC2277571 2.01 PYGO2 1.97 ZCCHC17 2.10 THC2448843 2.05 A_32_P166152 2.01 LOC389289 1.97 METTL4 2.10 BDP1 2.05 UTRN 2.01 PAOX 1.97 C20orf23 2.10 AK056744 2.05 HERC2 2.01 ING3 1.97 AGBL2 2.09 KLK1 2.05 THC2440422 2.01 BCOR 1.97 THC2281591 2.09 A_24_P255836 2.05 MOP-1 2.01 THC2341675 1.97 A_24_P832737 2.09 WDR26 2.05 AK022268 2.01 SPSB4 1.97 RP4-756G23.1 2.09 FLJ11235 2.05 A_24_P229903 2.01 GOT1 1.97 ELAC1 2.09 A_24_P383802 2.05 A_24_P169855 2.01 DDX46 1.97 CYP4V2 2.09 THC2441234 2.05 C6orf85 2.00 BC010544 1.97 LOC91137 2.09 IMPAD1 2.05 ENST00000366862 2.00 CR620804 1.97 BHLHB9 2.09 A_24_P289573 2.04 ZNF697 2.00 USP32 1.97 THC2404912 2.09 AP3B1 2.04 ENST00000370395 2.00 CYP1B1 1.97 ZFYVE1 2.08 RCHY1 2.04 F8 2.00 ZMYM5 1.97 GGTLA4 2.08 THC2435239 2.04 COQ5 2.00 TRIM62 1.96 KIAA0586 2.08 LOC284323 2.04 ENST00000358583 2.00 COL19A1 1.96 LRRC35 2.08 CDC25C 2.04 MLYCD 2.00 THC2336837 1.96 SASH1 2.08 A_24_P290114 2.04 USP6 2.00 TFR2 1.96 NDRG4 2.08 A_24_P375132 2.04 A_24_P367399 2.00 C6orf89 1.96 THC2435128 2.08 AK055306 2.04 ALPK2 2.00 LCAT 1.96 KLK2 2.08 AK056182 2.04 ENST00000374189 2.00 LQK1 1.96 KRT32 2.08 CD59 2.04 TAOK1 2.00 RASD1 1.96 A_23_P250072 2.08 VPS37B 2.04 ENST00000367846 2.00 C8orf6 1.96 PDK2 2.08 GALNT11 2.04 RSAD2 2.00 SMUG1 1.95 MLLT4 2.08 PPP2CB 2.03 ATG7 2.00 BC037919 1.95 VPS13D 2.08 HIVEP1 2.03 MYH10 2.00 CNOT7 1.95 C6orf27 2.08 ZNF284 2.03 THC2363295 1.99 C5orf4 1.95 AF336795 2.08 ENST00000266712 2.03 C9orf95 1.99 ZC3H5 1.95 GBP3 2.08 A_24_P384239 2.03 ADHFE1 1.99 KMO 1.95 CR590573 2.08 PNMA3 2.03 A_24_P25063 1.99 CRLF3 1.95 A_24_P135551 2.08 BU173515 2.03 C1orf82 1.99 GIMAP8 1.95 ENST00000372569 2.08 A_24_P15083 2.03 PJCG6 1.99 D4ST1 1.95 AF075028 2.08 AK074614 2.03 UBR2 1.99 KNS2 1.95 REV3L 2.08 LOC162073 2.03 THC2439430 1.99 MTA2 1.95 ZFAND5 2.08 CYBASC3 2.03 A_24_P375405 1.99 MTERFD2 1.94 PPFIBP1 2.07 ENST00000379734 2.03 GNG11 1.99 NOD9 1.94 MTERF 2.07 DND1 2.03 KCNK15 1.98 THC2311218 1.94 FUNDC1 2.07 REXO2 2.03 ETNK1 1.98 GATS 1.94 AK055501 2.07 AY010113 2.03 ZNF161 1.98 BI759100 1.94 FBN1 2.07 RMND5B 2.03 KLF12 1.98 THC2337268 1.94 A_24_P298604 2.07 A_32_P127412 2.03 PRO2266 1.98 FAM62B 1.94 ENST00000356126 2.07 THC2434618 2.03 THC2265846 1.98 AK124576 1.94 OTOF 2.07 VPS13A 2.02 ENST00000297423 1.98 THC2400529 1.94 RBM18 2.07 AK054939 2.02 ITPKB 1.98 SOCS1 1.94 CD86 2.07 SPTLC1 2.02 AK124281 1.98 C14orf101 1.94 A_23_P113453 2.07 MGC15523 2.02 COMMD6 1.98 THC2445411 1.94 ZNF167 2.07 ABHD9 2.02 THC2281903 1.98 A_24_P761386 1.94

132

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2314) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) MSH6 -7.24 LRRC20 -3.82 GPR89A -3.51 KCTD12 -3.28 CDC6 -6.91 FAM80A -3.80 IMPDH1 -3.50 FLJ14346 -3.28 UNG -6.46 SYMPK -3.80 SERPINH1 -3.50 HDAC7A -3.26 DTL -5.76 MANEAL -3.80 AK123655 -3.50 USP22 -3.26 CDCA7L -5.62 PYCR1 -3.79 EPN1 -3.50 C14orf153 -3.25 MCM4 -5.46 ZMYND19 -3.79 TM9SF4 -3.49 PPAN -3.25 FEN1 -5.28 A_24_P927072 -3.79 CR604497 -3.49 AP2A1 -3.24 ENST00000375256 -5.24 EIF2S2 -3.79 C16orf57 -3.49 A_24_P273014 -3.24 CDCA7 -5.19 PAXIP1 -3.78 ORC1L -3.48 ANKMY1 -3.24 A_24_P195400 -5.16 ERO1L -3.77 FSCN1 -3.48 GINS3 -3.24 CHAF1B -5.07 RASSF1 -3.77 NUDCD3 -3.46 RAD51 -3.24 BC079833 -4.97 FLJ30596 -3.76 MESP1 -3.45 TCOF1 -3.24 RBM21 -4.94 ARL6IP6 -3.76 A_24_P810074 -3.44 LARP2 -3.23 GNG7 -4.87 TUBGCP3 -3.75 FLJ40542 -3.43 BC022233 -3.23 UHRF1 -4.87 GPR146 -3.75 PFKFB4 -3.42 THC2313920 -3.23 CDC25A -4.70 ASS -3.75 A_24_P565908 -3.42 PDK1 -3.23 ETS2 -4.57 ENST00000311630 -3.75 CHRNA5 -3.42 PPAT -3.23 ZNF395 -4.55 C20orf20 -3.74 INSIG2 -3.41 ENST00000332107 -3.22 NCKIPSD -4.54 FKBP4 -3.73 DUSP7 -3.41 C1orf128 -3.22 WDR76 -4.53 MRPL17 -3.73 AYTL2 -3.41 AF132203 -3.21 A_24_P375962 -4.43 SREBF1 -3.73 TMUB1 -3.40 YEATS4 -3.21 CCNE2 -4.39 SPFH1 -3.72 A_24_P118721 -3.40 BC064349 -3.20 MAZ -4.34 RHOBTB2 -3.71 CDT1 -3.40 TRIM65 -3.20 GLOXD1 -4.28 SKP2 -3.71 TMEM121 -3.39 RFC2 -3.20 EXO1 -4.28 PIGW -3.70 DEGS1 -3.39 WWOX -3.20 TIMELESS -4.27 P2RY11 -3.70 SLFN11 -3.39 FXR2 -3.20 SLC1A5 -4.26 FLJ39779 -3.69 CENTB1 -3.39 LIMD1 -3.20 THC2334547 -4.24 AKT1 -3.68 CAMK1D -3.39 BRD2 -3.20 HNRPA0 -4.17 D21S2056E -3.68 TEAD4 -3.39 NBL1 -3.20 BE816002 -4.15 TFDP3 -3.68 SOX12 -3.38 HES6 -3.19 ING5 -4.13 A_24_P50328 -3.68 LOC388114 -3.38 ATAD2 -3.19 PPP5C -4.12 PFKFB3 -3.66 CCDC85B -3.38 CR596712 -3.19 GINS2 -4.11 LOC644063 -3.66 MYOHD1 -3.38 PPARA -3.19 FKBP5 -4.10 TCF3 -3.66 ING2 -3.38 BNIP3L -3.18 CR613972 -4.10 RCC1 -3.64 THC2404842 -3.37 ENST00000344035 -3.18 GIT1 -4.10 CD96 -3.63 PDXP -3.36 GEMIN4 -3.18 ORC6L -4.10 ITPKA -3.63 CD300C -3.36 STS -3.17 KLHL23 -4.08 OXCT2 -3.62 TNKS1BP1 -3.36 ZNF414 -3.16 MCM10 -4.07 C10orf119 -3.61 A_24_P221601 -3.35 DCPS -3.16 ENST00000382990 -4.07 CEBPE -3.61 KIAA1285 -3.35 ACY1L2 -3.16 ZFAND2B -4.06 RFC4 -3.60 ENST00000308269 -3.35 GALNTL4 -3.16 POLD3 -4.06 PAQR4 -3.59 SUV420H2 -3.35 CENPQ -3.16 ENST00000373316 -4.06 A_24_P714707 -3.58 JAKMIP1 -3.34 THC2407148 -3.15 ENO2 -4.02 AK022183 -3.58 MAD2L1 -3.34 CAMKK1 -3.15 ENST00000327299 -4.01 NUPL1 -3.57 RNPS1 -3.33 ENST00000307214 -3.14 POLD1 -4.00 HLA-G -3.57 UBA2 -3.32 ATPBD3 -3.14 SLC8A3 -3.99 CR625518 -3.57 ST3GAL1 -3.32 BC087732 -3.14 NUP205 -3.97 C14orf24 -3.56 SLC43A1 -3.31 ZNF146 -3.14 KIAA1715 -3.93 ST6GALNAC6 -3.56 HNRPDL -3.31 SLC39A3 -3.13 FUT11 -3.92 C22orf13 -3.56 DBP -3.31 CHRAC1 -3.13 MTAP -3.91 AV749257 -3.56 PGEA1 -3.31 NR2F6 -3.13 LOC162427 -3.91 SLC19A1 -3.54 L07392 -3.31 ELK1 -3.13 MINK1 -3.90 NANP -3.54 AF119906 -3.30 WDR79 -3.12 ZADH2 -3.90 GRAP -3.53 ATAD3B -3.29 THC2363476 -3.12 RPP25 -3.90 PIK3R2 -3.53 C9orf32 -3.29 FLJ35773 -3.11 TMEM109 -3.89 BU618765 -3.53 DTX1 -3.29 LARP1 -3.11 SPHK1 -3.86 SLC1A1 -3.53 SETD1A -3.28 HCFC1R1 -3.11 C5orf25 -3.86 PDLIM7 -3.52 KIAA1919 -3.28 A_24_P92411 -3.11 TNF -3.83 SLC31A1 -3.52 PSMD11 -3.28 SCD -3.11 ARHGEF1 -3.83 BCAT1 -3.51 AGPAT2 -3.28 MLLT7 -3.10

133

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2314) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) A_24_P383751 -3.10 KIAA1754L -2.96 THC2320434 -2.86 H1FX -2.78 POU3F1 -3.10 CDCA4 -2.96 A_24_P233560 -2.86 SUV39H1 -2.78 GLT25D1 -3.10 GGA2 -2.96 CCDC19 -2.85 NIPSNAP1 -2.78 POLR3K -3.10 ENSA -2.96 MIPEP -2.85 CHEK1 -2.78 C14orf130 -3.10 GANAB -2.96 LOC133619 -2.85 TMEM44 -2.78 CHMP6 -3.10 SH2D5 -2.95 LRRC42 -2.85 IQCC -2.78 VLDLR -3.10 VAT1 -2.95 BRD9 -2.85 TOM1 -2.77 LOC146439 -3.09 LOC442013 -2.95 RNF126 -2.85 UAP1L1 -2.77 CELSR3 -3.09 B4GALT2 -2.95 TNFRSF12A -2.84 TMTC4 -2.77 TBL2 -3.09 AW467174 -2.95 FASN -2.84 C18orf24 -2.77 ATXN7L3 -3.09 A_24_P255954 -2.95 A_24_P834646 -2.84 ZNF343 -2.77 PPME1 -3.09 AXIN1 -2.94 BC092452 -2.84 METAP1 -2.76 ENST00000380635 -3.09 SCMH1 -2.94 RASSF7 -2.84 CHDH -2.76 LOC643668 -3.08 TPCN1 -2.94 PRIM1 -2.84 GBL -2.76 TAGLN2 -3.08 PER1 -2.94 A_32_P108420 -2.84 PTPN7 -2.76 ING1 -3.08 TRIT1 -2.94 ASB13 -2.84 CCDC71 -2.76 BARD1 -3.08 RAB33A -2.94 RP3-402G11.5 -2.83 AK091337 -2.76 CENTG3 -3.08 ZC3H10 -2.94 A_24_P247616 -2.83 A_24_P187304 -2.76 PUS7 -3.08 HES4 -2.94 VEGFB -2.83 ABHD11 -2.76 XRCC3 -3.08 STK24 -2.93 THC2305336 -2.83 A_24_P24453 -2.76 CNOT3 -3.08 AK023376 -2.93 ZC3H14 -2.83 UNC84A -2.76 CLSPN -3.08 NOC2L -2.92 FLJ40722 -2.83 ABL1 -2.76 HSD17B6 -3.07 CENTA1 -2.92 GART -2.83 KIAA1509 -2.75 BAT4 -3.07 DDX55 -2.92 VPS26B -2.82 PTGES2 -2.75 N4BP3 -3.06 MAPK9 -2.92 TOMM40 -2.82 THADA -2.75 PTPLAD1 -3.06 MVK -2.92 AA974271 -2.82 MTHFD1L -2.75 CAMKK2 -3.05 BI050742 -2.92 LENG1 -2.82 ASNA1 -2.75 CDC45L -3.05 AES -2.92 KRT18 -2.82 MMP25 -2.75 XPOT -3.05 RTF1 -2.92 GDI2 -2.82 NUP88 -2.75 CHMP7 -3.05 SCAP1 -2.92 TJP2 -2.82 A_24_P203976 -2.75 TP73 -3.04 EHMT2 -2.92 BC047380 -2.81 PES1 -2.74 MSH2 -3.04 RNF157 -2.92 AL522622 -2.81 A_24_P594094 -2.74 CACNB3 -3.04 G0S2 -2.91 ENST00000370290 -2.81 STARD3 -2.74 CD3EAP -3.04 PTK2B -2.91 UBE2MP1 -2.81 FOXJ2 -2.73 LOC339344 -3.04 ARFRP1 -2.91 MGC13005 -2.81 STK35 -2.73 CSNK2A1 -3.03 RAB1B -2.91 ELMO2 -2.81 PHF19 -2.73 VAMP2 -3.03 OMP -2.91 NEIL3 -2.81 CR596214 -2.73 BC031876 -3.03 TMEM86B -2.91 hCAP-H2 -2.81 TRAIP -2.73 RP1-93H18.5 -3.02 CIRH1A -2.90 GMPPB -2.81 FLNA -2.73 SLC39A8 -3.02 MAD2L2 -2.90 TGIF2 -2.80 BE003490 -2.72 ENST00000312785 -3.02 SLC6A6 -2.90 COX19 -2.80 SPAG9 -2.72 ENST00000375590 -3.01 DDN -2.90 MAPK14 -2.80 CXorf15 -2.72 ENST00000382579 -3.01 RGC32 -2.90 BC036361 -2.80 PEX5 -2.72 BAG5 -3.01 RFC5 -2.90 SAMD1 -2.80 A_24_P470809 -2.72 GPR157 -3.01 A_24_P358302 -2.90 MGC4268 -2.80 SLC25A19 -2.72 BM932296 -3.00 A_24_P692600 -2.89 HNRPD -2.80 ABCB8 -2.72 A_24_P101742 -2.99 WNT10A -2.89 A_24_P255845 -2.79 COPA -2.72 CDK2 -2.99 ANKRD40 -2.89 AF118084 -2.79 CDC7 -2.72 ZNF23 -2.99 STIP1 -2.89 ZNF473 -2.79 ENST00000288548 -2.72 KLHDC3 -2.99 RASA4 -2.89 ZNF687 -2.79 SIAH1 -2.72 THC2268216 -2.99 KHK -2.88 BQ072652 -2.79 A_23_P88554 -2.72 LSM12 -2.99 VCP -2.88 ENST00000359653 -2.79 A_24_P409816 -2.72 MLLT4 -2.99 FKBPL -2.88 HSP90AB3P -2.78 ST3GAL3 -2.72 TPD52L2 -2.98 ENST00000369158 -2.88 A_24_P170309 -2.78 A_24_P771278 -2.71 MGC4562 -2.98 SFRS4 -2.88 RAB8A -2.78 BC063441 -2.71 A_24_P539275 -2.97 LRRC45 -2.87 WBSCR1 -2.78 EPS8L1 -2.71 ADCK2 -2.97 POLE -2.87 SLC16A1 -2.78 RASGRP1 -2.71 ANKRD25 -2.97 C12orf41 -2.86 A_24_P178475 -2.78 NDE1 -2.71 EXOC6 -2.97 ANKRD41 -2.86 AK092163 -2.78 KCNAB2 -2.71 CAMK1 -2.96 PTP4A2 -2.86 MAPK3 -2.78 CHKA -2.71

134

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2314) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) HSF2 -2.71 BTBD14B -2.64 ANP32B -2.59 CTSC -2.52 ENST00000332534 -2.71 UPP1 -2.64 LOC339123 -2.58 AFMID -2.52 PGRMC2 -2.71 PQLC1 -2.64 LOC642711 -2.58 GORASP2 -2.51 THC2407189 -2.71 OGG1 -2.64 LOC399491 -2.58 TMEM93 -2.51 SCLY -2.71 C16orf55 -2.64 CXorf9 -2.58 LOC440895 -2.51 ENST00000318251 -2.70 ARHGDIA -2.64 PUM2 -2.58 CHST10 -2.51 TRSPAP1 -2.70 UHRF2 -2.64 ETV4 -2.58 DDX42 -2.51 KIAA1794 -2.70 BQ337821 -2.64 SLC25A23 -2.57 GNB5 -2.51 A_24_P170283 -2.70 ZRANB1 -2.63 C8orf55 -2.57 CCDC28B -2.51 USP28 -2.70 C10orf95 -2.63 ODF2 -2.57 RP11-11C5.2 -2.51 RHEB -2.70 NOP17 -2.63 WBP2 -2.57 KLHL17 -2.51 BRF1 -2.70 A_24_P332721 -2.63 TETRAN -2.57 RSL1D1 -2.50 RBM38 -2.70 GLUD2 -2.63 PHF13 -2.57 C20orf11 -2.50 TSR2 -2.69 RIPK2 -2.63 A_24_P530690 -2.56 NFKBIL1 -2.50 ADCK4 -2.69 PIK3R5 -2.63 TMEM18 -2.56 MBD3 -2.50 PDDC1 -2.69 ENOSF1 -2.63 CUTC -2.56 PTCD2 -2.50 ANP32D -2.69 A_24_P84698 -2.63 THC2376568 -2.56 GTPBP1 -2.50 TLOC1 -2.69 C4orf9 -2.63 AK026497 -2.56 ERICH1 -2.50 A_24_P418498 -2.69 MSN -2.63 A_24_P289504 -2.56 RARG -2.50 ENST00000340455 -2.69 PRR6 -2.63 GSPT1 -2.56 FZR1 -2.50 CCDC86 -2.69 THC2273298 -2.62 LYAR -2.56 FLJ11021 -2.50 WDR46 -2.69 KHSRP -2.62 PWP2H -2.56 NAGPA -2.50 KDELC2 -2.69 BC031940 -2.62 USP10 -2.56 MAPK12 -2.50 CEBPB -2.69 PELP1 -2.62 MGC16703 -2.56 BAT1 -2.49 AFG3L2 -2.68 HMG20B -2.62 KIAA0133 -2.55 C14orf172 -2.49 YWHAE -2.68 IL16 -2.62 RGS14 -2.55 03/01/2009 -2.49 UIP1 -2.68 ZNF57 -2.62 ZNF507 -2.55 TRIB3 -2.49 BF887921 -2.68 FLJ12529 -2.62 WDHD1 -2.55 P4HA1 -2.49 AKT1S1 -2.68 METT5D1 -2.62 A_23_P251196 -2.55 A_23_P75129 -2.49 A_32_P46700 -2.68 PKN1 -2.62 THC2280176 -2.55 CHCHD4 -2.49 PKP4 -2.68 UNC5CL -2.62 DNA2L -2.55 NT5DC3 -2.49 ST6GAL1 -2.67 SIT1 -2.62 AA495894 -2.55 FLJ10154 -2.49 MGC40405 -2.67 A_24_P221485 -2.62 BICD2 -2.55 ABCC1 -2.48 ZNF651 -2.67 PTGES3 -2.62 ENST00000367142 -2.55 BC047717 -2.48 CTA-126B4.3 -2.67 SFI1 -2.62 B3GAT3 -2.54 PYCRL -2.48 LAT -2.67 RPRC1 -2.61 LYPLA2 -2.54 BE855644 -2.48 ACSS1 -2.67 VEGF -2.61 HSP90AB1 -2.54 GAS8 -2.48 DCC1 -2.67 LOC441320 -2.61 MGC11271 -2.54 KIAA0241 -2.48 76P -2.67 SLC7A5 -2.61 ALKBH5 -2.54 CR603272 -2.48 GLI4 -2.67 MGC42157 -2.61 DONSON -2.54 SLC43A3 -2.48 C17orf58 -2.67 C10orf137 -2.61 THC2357608 -2.54 HERC2 -2.48 DEPDC4 -2.66 THAP7 -2.61 GINS1 -2.54 TICAM1 -2.48 GPIAP1 -2.66 C16orf75 -2.61 A_24_P367397 -2.54 AMFR -2.48 TOE1 -2.66 NUP93 -2.61 CD151 -2.53 CCDC49 -2.48 NCL -2.66 THC2263651 -2.60 DNMT1 -2.53 RHBDL3 -2.48 GTL3 -2.66 MPDU1 -2.60 C10orf9 -2.53 DHCR24 -2.48 REPS1 -2.66 ENST00000315293 -2.60 AP3M2 -2.53 AK126814 -2.48 VPRBP -2.66 EGR1 -2.60 ATP11B -2.53 LOC643431 -2.48 NUFIP1 -2.66 ZNF230 -2.60 GTPBP3 -2.53 ENST00000359244 -2.47 ADD1 -2.66 EGLN1 -2.60 CR609948 -2.53 ZC3HAV1L -2.47 HYAL2 -2.66 CD300A -2.60 C14orf65 -2.53 HES5 -2.47 ENO1B -2.66 SPAG4 -2.60 GMPS -2.53 CENPB -2.47 BTN3A1 -2.66 C20orf59 -2.60 SETD3 -2.53 LOC90355 -2.47 KIAA0963 -2.65 SLAMF6 -2.60 TMED2 -2.53 CTCF -2.47 ZFP161 -2.65 ZNF530 -2.60 E2F1 -2.53 NT5M -2.47 AK3L2 -2.65 ITGB1BP1 -2.59 GMEB2 -2.52 MBD1 -2.47 HDGF -2.65 PRKCSH -2.59 C22orf9 -2.52 SFRS1 -2.47 KIAA1542 -2.65 C11orf75 -2.59 RAC3 -2.52 LENG9 -2.47 UBL4A -2.65 DEK -2.59 KLHL22 -2.52 FUZ -2.47 LOC150223 -2.64 A_24_P92823 -2.59 ARHGEF2 -2.52 BRI3BP -2.47

135

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2314) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) CARHSP1 -2.47 USP5 -2.42 ANKZF1 -2.36 CHAF1A -2.30 PTPMT1 -2.47 KDELR1 -2.41 ACVR2B -2.36 SCNN1B -2.30 EML3 -2.46 MAFF -2.41 THC2405842 -2.36 SBDS -2.30 NOLC1 -2.46 ACADS -2.41 LOC196394 -2.36 GSTCD -2.30 KIAA0664 -2.46 A_24_P391853 -2.41 SOX7 -2.35 RHOF -2.30 FBXO5 -2.46 T05215 -2.41 THC2315472 -2.35 MGC23280 -2.30 PHF20 -2.46 AK057198 -2.41 SET -2.35 LMAN1 -2.30 MRPL43 -2.46 AK025344 -2.41 CD320 -2.35 C2orf15 -2.30 A_24_P50281 -2.46 DFFB -2.41 DPP9 -2.35 FOXK2 -2.30 ARL4D -2.46 ABTB1 -2.41 ZBTB6 -2.35 CBX6 -2.30 ELF4 -2.46 BCL2L13 -2.41 ANAPC5 -2.35 A_24_P796274 -2.29 METTL1 -2.46 CC2D1A -2.40 MYBL2 -2.35 IGHMBP2 -2.29 ASXL1 -2.45 IL12RB2 -2.40 UBTF -2.35 YARS -2.29 RBM15B -2.45 LAT2 -2.40 FAM57A -2.34 LOC390705 -2.29 EXDL2 -2.45 PCOLN3 -2.40 DNAJB2 -2.34 FZD1 -2.29 FARSLB -2.45 LOC389458 -2.40 PBEF1 -2.34 SLC2A1 -2.29 PLCH2 -2.45 GIPC1 -2.40 DCLRE1B -2.34 A_32_P114146 -2.29 CRKL -2.45 AF119911 -2.40 BC047032 -2.34 AK125393 -2.29 ATAD1 -2.45 PLCXD1 -2.40 PCP2 -2.34 SYNJ2 -2.29 FBXO36 -2.45 PIGA -2.40 UCN -2.34 C14orf43 -2.29 POLR1A -2.45 TMEM16F -2.40 LY9 -2.34 DA234975 -2.29 C14orf122 -2.45 A_32_P219704 -2.39 YOD1 -2.34 MUM1 -2.29 THUMPD3 -2.45 AK054718 -2.39 CACYBP -2.34 BX346853 -2.29 HABP4 -2.45 MMAB -2.39 THC2347074 -2.34 BX094072 -2.29 CKAP4 -2.45 ST7 -2.39 MFSD2 -2.34 ERGIC1 -2.28 CLK2 -2.45 DDX11 -2.39 ANP32A -2.34 DAGLBETA -2.28 UAP1 -2.44 A_24_P50294 -2.39 C6orf96 -2.33 A_24_P178154 -2.28 USF2 -2.44 TRIP13 -2.39 AK093548 -2.33 ENST00000304245 -2.28 TRA16 -2.44 EIF2S1 -2.39 OPN3 -2.33 AK131288 -2.28 STT3A -2.44 A_24_P324214 -2.39 TNRC5 -2.33 BXDC2 -2.28 TSR1 -2.44 MLSTD1 -2.39 RNF6 -2.33 FMNL2 -2.28 CLCN2 -2.44 RAB5C -2.38 ENST00000331406 -2.33 TLE3 -2.28 SHQ1 -2.44 FIGNL1 -2.38 AK021629 -2.33 APBB1 -2.28 A_32_P152696 -2.44 KIAA1008 -2.38 MGC16037 -2.33 C20orf43 -2.28 YTHDF1 -2.44 APOE -2.38 PIGX -2.33 NOB1 -2.28 THRAP3 -2.44 SR-A1 -2.38 AF103312 -2.33 NUDT4 -2.28 NELF -2.44 SYTL3 -2.38 XBP1 -2.33 KCTD20 -2.28 C19orf47 -2.44 C1orf33 -2.38 EEF2K -2.33 THC2374166 -2.27 SIAH2 -2.44 SYNC1 -2.37 C9orf140 -2.33 ITGAL -2.27 A4GALT -2.43 GNAZ -2.37 RPL7L1 -2.32 FJX1 -2.27 GFPT1 -2.43 MLLT6 -2.37 PRPS2 -2.32 SMARCB1 -2.27 NR1D2 -2.43 ITPR3 -2.37 VKORC1 -2.32 LRP8 -2.27 SERGEF -2.43 PPP2R5B -2.37 THC2409354 -2.32 ADAM15 -2.27 A_24_P101181 -2.43 A_24_P213336 -2.37 FADS2 -2.32 SMO -2.27 SMYD4 -2.43 EIF2C2 -2.37 BX457454 -2.32 FH -2.27 FAM86C -2.43 RNF167 -2.37 MTMR12 -2.32 A_24_P455100 -2.27 PEX16 -2.43 CX164944 -2.37 A_24_P835943 -2.32 APOL1 -2.27 C19orf28 -2.43 MYBBP1A -2.37 ATP6V1E2 -2.31 GMCL1 -2.27 D89937 -2.43 AP1S1 -2.37 BE467780 -2.31 A_24_P401150 -2.27 APH1A -2.43 SURF5 -2.37 A_24_P853302 -2.31 PPRC1 -2.27 dJ222E13.2 -2.42 MGC2408 -2.37 SMN2 -2.31 A_24_P24806 -2.27 ZNF598 -2.42 RNASEH2A -2.37 WDR25 -2.31 MGC4655 -2.27 TCP10L -2.42 MAX -2.37 SFRS15 -2.31 TCP1 -2.27 BAT2 -2.42 TIGD3 -2.36 AK093691 -2.31 USP37 -2.27 SUV39H2 -2.42 ZNF263 -2.36 KLHL21 -2.31 DNAL4 -2.27 NOS3 -2.42 ERN1 -2.36 C9orf41 -2.31 A_24_P625898 -2.27 EFCAB4B -2.42 ABCG1 -2.36 GNAI2 -2.31 ST3GAL4 -2.27 M-RIP -2.42 E2F2 -2.36 BSPRY -2.31 REXO4 -2.26 ESAM -2.42 A_24_P213256 -2.36 A_24_P101503 -2.30 ENST00000381800 -2.26 HNRPAB -2.42 RFX2 -2.36 ENST00000369731 -2.30 A_24_P349580 -2.26

136

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2314) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) TNPO1 -2.26 MGC11257 -2.23 KIAA0831 -2.19 LOC126208 -2.15 AL526449 -2.26 AA843546 -2.23 LOC375133 -2.19 TNPO2 -2.15 NRF1 -2.26 PDIA3 -2.22 ENST00000220507 -2.19 A_24_P534290 -2.15 A_24_P126851 -2.26 IFT52 -2.22 A_24_P152278 -2.19 RBM23 -2.15 TPD52 -2.26 DGAT2 -2.22 MYO1C -2.19 MAT2A -2.15 NARS2 -2.26 PFKL -2.22 TRERF1 -2.18 SAAL1 -2.15 CDK10 -2.26 A_23_P31563 -2.22 ITGA2B -2.18 DKFZp761E198 -2.15 A_24_P341489 -2.26 GTPBP6 -2.22 PSMC3IP -2.18 HM13 -2.15 HPCA -2.26 SIX5 -2.22 ZNF136 -2.18 A_24_P306614 -2.15 DVL1 -2.26 HEBP2 -2.22 CCND3 -2.18 KIAA1704 -2.15 CGI-09 -2.26 SNX17 -2.22 TMEM125 -2.18 PPM1G -2.15 A_24_P401124 -2.26 TTC19 -2.22 TRIM26 -2.18 BF246504 -2.15 POLE2 -2.26 MTERFD1 -2.22 RNF149 -2.18 A_32_P92922 -2.15 VGLL4 -2.26 NAB2 -2.22 COMTD1 -2.18 SMPD4 -2.14 C6orf64 -2.26 ILKAP -2.22 POLD2 -2.18 IFRD2 -2.14 09/01/2006 -2.25 AI675560 -2.22 EPB41L2 -2.18 LIG3 -2.14 MIDN -2.25 C6orf61 -2.22 ZFP41 -2.18 C2orf7 -2.14 THC2270231 -2.25 C6orf72 -2.21 A_24_P169574 -2.18 GRPEL2 -2.14 PPIL5 -2.25 DDX31 -2.21 RAB43 -2.18 LRP2BP -2.14 EMID1 -2.25 KATNB1 -2.21 C8orf33 -2.18 CSRP1 -2.14 TOMM34 -2.25 AI354226 -2.21 A_23_P99731 -2.18 TTL -2.14 C3orf21 -2.25 PTDSR -2.21 PLOD1 -2.18 CTNNAL1 -2.14 NUP155 -2.25 A_24_P306704 -2.21 GRLF1 -2.17 FANCE -2.14 EXOSC8 -2.25 FOXP4 -2.21 LOC152719 -2.17 NDST1 -2.14 RAI1 -2.25 BF869497 -2.21 COTL1 -2.17 ALG9 -2.14 AMPD2 -2.25 HDGF2 -2.21 A_24_P16071 -2.17 A_24_P383680 -2.14 AGPAT6 -2.25 A_24_P247493 -2.21 ZNF394 -2.17 FEM1C -2.14 TRAFD1 -2.25 KCNH3 -2.21 BQ310837 -2.17 TERT -2.14 TAZ -2.25 CXorf24 -2.21 GMNN -2.17 NME4 -2.14 UBQLN4 -2.25 WDR42A -2.21 HDAC4 -2.17 THC2430394 -2.14 ZBTB2 -2.25 PLEKHF2 -2.20 C19orf24 -2.17 PIP5K1A -2.14 C3orf26 -2.25 ARHGAP27 -2.20 XPO6 -2.17 FLJ38984 -2.14 RAB11FIP1 -2.24 THC2340379 -2.20 MYBPH -2.17 BC051742 -2.14 THC2311196 -2.24 UCK1 -2.20 EFEMP2 -2.17 A_24_P626812 -2.14 SLC27A4 -2.24 ALG3 -2.20 ENST00000095634 -2.17 PPOX -2.14 DTX3 -2.24 MUTED -2.20 HIST1H1B -2.17 LIMK2 -2.13 ATP2A3 -2.24 BC019599 -2.20 A_24_P307424 -2.17 BQ008507 -2.13 BM984383 -2.24 ENST00000357180 -2.20 ARSG -2.17 PRIM2A -2.13 THOC6 -2.24 A_24_P255865 -2.20 PPT1 -2.16 A_24_P349648 -2.13 AARS -2.24 A_24_P612921 -2.20 TAF1A -2.16 A_24_P67574 -2.13 PACSIN1 -2.24 DHFR -2.20 A_24_P375510 -2.16 URP2 -2.13 FLJ20245 -2.24 SCRN1 -2.20 LOC144097 -2.16 NUP210 -2.13 HNRPCL1 -2.24 RIPK4 -2.20 PPIE -2.16 SNRP70 -2.13 CAMTA2 -2.24 EIF4EBP2 -2.20 JAK3 -2.16 ENST00000269142 -2.13 SHMT2 -2.24 TFDP1 -2.20 ZNF232 -2.16 ACIN1 -2.13 ABCA2 -2.24 CEACAM21 -2.20 PMM1 -2.16 SH3GL1 -2.13 LDLR -2.24 GNL1 -2.20 ZFP36 -2.16 JUNB -2.13 LEF1 -2.24 MASTL -2.20 A_24_P862251 -2.16 TMEM19 -2.13 ZNHIT4 -2.24 C18orf17 -2.19 FAM120C -2.16 C9orf75 -2.13 A_32_P6274 -2.23 A_24_P323916 -2.19 ADPRHL2 -2.16 LGALS8 -2.12 C15orf39 -2.23 MTP18 -2.19 KIAA0738 -2.16 KIAA1430 -2.12 A_23_P300563 -2.23 SLC1A4 -2.19 POLR2E -2.16 GCN5L2 -2.12 ARL3 -2.23 OTUD1 -2.19 S100PBP -2.16 TXLNA -2.12 BC011660 -2.23 ZNF331 -2.19 A_24_P835388 -2.16 CRSP7 -2.12 KIAA2013 -2.23 SLC20A1 -2.19 A_24_P200962 -2.16 HIST1H1E -2.12 MUTYH -2.23 SHMT1 -2.19 A_24_P187154 -2.16 ENST00000328046 -2.12 THC2369600 -2.23 C1orf135 -2.19 HAGHL -2.16 PRR3 -2.12 THC2276742 -2.23 THC2430400 -2.19 C3orf40 -2.16 A_24_P84873 -2.12 ZNF160 -2.23 HK2 -2.19 CXorf34 -2.16 AK026811 -2.12 CPSF3L -2.23 TNFRSF13B -2.19 ENST00000360741 -2.16 A_24_P195621 -2.12

137

TABLE S-T7: BRCA2 (+/-) IR-RESPONSE DOWN-REGULATED GENES (2314) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) Gene ID Score(d) BCAT2 -2.12 APC -2.08 KIAA1815 -2.04 SFXN1 -2.01 EML2 -2.12 SEMA4D -2.08 HEG1 -2.04 RBM13 -2.01 LOC645249 -2.11 CHD8 -2.08 DUSP16 -2.04 PKMYT1 -2.01 A_24_P477102 -2.11 RUNDC1 -2.08 LOC203427 -2.04 PPARD -2.01 KIAA0683 -2.11 SNRPD1 -2.08 C3orf39 -2.04 LAMB3 -2.01 PRH2 -2.11 A_24_P324250 -2.08 PIGC -2.04 A_32_P143530 -2.01 SEPHS1 -2.11 THBS4 -2.08 CCDC107 -2.04 BRCA2 -2.01 U2AF2 -2.11 THC2316236 -2.08 FEM1A -2.04 GTF3C2 -2.00 ENST00000358396 -2.11 TXNDC12 -2.08 SLC3A2 -2.04 EIF4G1 -2.00 THC2403745 -2.11 LOC442075 -2.08 MSTO1 -2.04 AI857589 -2.00 C6orf167 -2.11 PPP1CB -2.08 SMARCC2 -2.04 RNF122 -2.00 C3orf15 -2.11 A_24_P143653 -2.08 CERK -2.04 CCDC117 -2.00 POU5F1 -2.11 A_24_P392099 -2.07 SUMO2 -2.04 MAPK8 -2.00 UNC93B1 -2.10 ST13 -2.07 FBXL14 -2.04 ARAF -2.00 P2RX3 -2.10 WDR21A -2.07 RAE1 -2.03 HTF9C -2.00 RUVBL1 -2.10 DOCK6 -2.07 CRY2 -2.03 ENST00000320415 -2.00 SLC12A9 -2.10 SUFU -2.07 FAM111B -2.03 EIF4EBP1 -2.00 OSR2 -2.10 STX11 -2.07 AK021676 -2.03 SPINT1 -2.00 A_24_P401392 -2.10 KIAA1862 -2.07 SPRED1 -2.03 A_32_P133926 -2.00 IVD -2.10 PTTG2 -2.07 ENST00000377703 -2.03 RAD51L3 -2.00 GRPEL1 -2.10 AHNAK -2.07 POLR1E -2.03 AA725860 -2.00 TP53 -2.10 TULP3 -2.07 ACTL6A -2.03 FKBP8 -2.00 ANKRD13D -2.10 STYX -2.07 BF689038 -2.03 NUDT9P1 -2.00 CB853344 -2.10 LONRF3 -2.07 THC2305868 -2.03 TMEM39B -2.00 C16orf59 -2.10 FAM81A -2.07 FAM89A -2.03 BM455859 -2.00 OS9 -2.10 LOC642378 -2.07 DGKE -2.03 A_24_P50390 -2.00 C9orf74 -2.10 TAF6L -2.06 AK054852 -2.03 NFKBIL2 -2.00 FIP1L1 -2.10 TBC1D9B -2.06 NFIC -2.03 NAT13 -2.00 C16orf74 -2.10 GLTSCR1 -2.06 ZNF84 -2.03 MTBP -2.00 FAM40B -2.10 C18orf19 -2.06 NY-SAR-48 -2.02 RAD23A -2.00 PTPN23 -2.10 LOC203547 -2.06 MTA1 -2.02 ENST00000323501 -1.99 HEMK1 -2.09 FADS1 -2.06 WHSC1 -2.02 USP7 -1.99 FAM119A -2.09 CCDC109A -2.06 LOC401010 -2.02 SDCBP2 -1.99 PRKAR1B -2.09 THC2363646 -2.06 TMEM102 -2.02 C9orf91 -1.99 CNTROB -2.09 FPGS -2.06 HNRPH1 -2.02 TM2D2 -1.99 PGS1 -2.09 GNPDA1 -2.06 SFT2D3 -2.02 RNF135 -1.99 HPCAL1 -2.09 OSBP -2.06 PARP16 -2.02 MAPK8IP1 -1.99 MLLT1 -2.09 CEBPG -2.06 A_24_P767725 -2.02 DBR1 -1.99 MOCS3 -2.09 ZBTB7C -2.06 RDH13 -2.02 A_24_P118813 -1.99 NASP -2.09 KBTBD2 -2.06 LOC286260 -2.02 COG3 -1.99 DPP7 -2.09 NONO -2.06 ZNF566 -2.02 HCP1 -1.99 KIAA0251 -2.09 AGPAT7 -2.06 DKC1 -2.02 THC2309960 -1.99 NP1165618 -2.09 CHID1 -2.06 CNDP2 -2.02 THC2304728 -1.99 ABHD12 -2.09 STAT5B -2.06 A_24_P221960 -2.02 FAM35A -1.99 LOC285074 -2.09 BTD -2.05 WDTC1 -2.02 A_24_P75708 -1.99 RIOK1 -2.09 CXXC5 -2.05 IDI1 -2.02 MAML1 -1.99 LSR -2.09 UBE2M -2.05 A_24_P846755 -2.02 GPR155 -1.99 C6orf32 -2.09 FLOT1 -2.05 GPSM2 -2.02 PHC2 -1.99 ZNF692 -2.09 LMAN2 -2.05 PUS1 -2.02 A_23_P10987 -1.99 SIPA1L1 -2.09 SIPA1 -2.05 ENST00000270238 -2.01 SPIB -1.99 FBXO11 -2.09 BTLA -2.05 MKNK2 -2.01 ANKS3 -1.98 CORO2A -2.09 AK094415 -2.05 AKT2 -2.01 SCYL1 -1.98 DOCK7 -2.09 TTC1 -2.05 IPO11 -2.01 A_24_P651129 -1.98 SYT15 -2.08 BC016872 -2.05 SGK3 -2.01 CDKN3 -1.98 MGC11102 -2.08 PARP11 -2.05 TXNRD2 -2.01 ENST00000309878 -1.98 ATP8B2 -2.08 A_24_P584463 -2.05 PAX5 -2.01 JOSD1 -1.98 A_24_P853004 -2.08 PHF17 -2.05 PTPN4 -2.01 THC2281305 -1.98 E2F4 -2.08 NRBP1 -2.05 NUMA1 -2.01 PHKA2 -1.98 TRAPPC2L -2.08 SCAND1 -2.05 SLC25A25 -2.01 RCC2 -1.98 AHSA1 -2.08 NTHL1 -2.05 CYB5R3 -2.01 LOC90321 -1.98

138