Novel Primary Immunodeficiency Candidate Genes Predicted by The

Home , GLMN, GRLF1, HSP90AA2, Phosphoglucomutase 3, PRKRIR

ORIGINAL RESEARCH ARTICLE published: 01 April 2015 doi: 10.3389/ﬁmmu.2015.00142 Novel primary immunodeﬁciency candidate genes predicted by the human gene connectome

Yuval Itan1* and Jean-Laurent Casanova1,2,3,4,5

1 Rockefeller Branch, St. Giles Laboratory of Human Genetics of Infectious Diseases, The Rockefeller University, New York, NY, USA 2 Necker Branch, Laboratory of Human Genetics of Infectious Diseases, INSERM U1163, Paris, France 3 Imagine Institute, University Paris Descartes, Paris, France 4 Howard Hughes Medical Institute, New York, NY, USA 5 Pediatric Hematology-Immunology Unit, Necker Hospital for Sick Children, Paris, France

Edited by: Germline genetic mutations underlie various primary immunodeficiency (PID) diseases. Fabio Candotti, CHUV – Centre Patients with rare PID diseases (like most non-PID patients and healthy individuals) carry, on Hospitalier Universitaire Vaudois, Switzerland average, 20,000 rare and common coding variants detected by high-throughput sequenc- Reviewed by: ing. It is thus a major challenge to select only a few candidate disease-causing variants Sergio Rosenzweig, National for experimental testing. One of the tools commonly used in the pipeline for estimating a Institutes of Health, USA potential PID-candidate gene is to test whether the specific gene is included in the list of Silvia Clara Giliani, University of genes that were already experimentally validated as PID-causing in previous studies. How- Brescia, Italy ever, this approach is limited because it cannot detect the PID-causing mutation(s) in the *Correspondence: Yuval Itan, St. Giles Laboratory of many PID patients carrying causal mutations of as yet unidentified PID-causing genes. In Human Genetics of Infectious this study, we expanded in silico the list of potential PID-causing candidate genes from 229 Diseases, The Rockefeller University, to 3,110. We first identified the top 1% of human genes predicted by the human genes New York, NY, USA connectome to be biologically close to the 229 known PID genes. We then further nar- e-mail: [email protected] rowed down the list of genes by retaining only the most biologically relevant genes, with functionally enriched gene ontology biological categories similar to those for the known PID genes. We validated this prediction by showing that 17 of the 21 novel PID genes published since the last IUIS classification fall into this group of 3,110 genes (p < 10−7).The resulting new extended list of 3,110 predicted PID genes should be useful for the discovery of novel PID genes in patients.

Keywords: candidate gene identiﬁcation, primary immunodeﬁciencies, the human gene connectome, disease genomics, computational biology

INTRODUCTION estimating potential biological functional relatedness to the known Germline mutations are being found to underlie an increasing PID genes. number of primary immunodeficiency (PID) diseases. With cur- To tackle the above problem, we recently described the human rent advances, and the improved quality and decreasing cost gene connectome (HGC)1 as the set of biological distances and of high-throughput sequencing (HTS), it is now possible to routes (i.e., genes located between two genes) between all human detect the full set of gene variants in PID patients, through genes, predicted in silico by a shortest distance algorithm applied techniques such as whole-exome sequencing (WES) or whole- to the full human genome network, conceptually similar to GPS genome sequencing (WGS). The genome of each patient con- navigation (5). The HGC and its associated user-friendly server tains about 20,000 coding variants and hundreds of thousands (8) could be used to detect new candidate PID-causing genes, by of non-coding variants (1–3). It is not straightforward to iden- ranking all genes harboring variants in PID patients on the basis of tify the PID-causing gene from the HTS data for a patient (4, their biological proximity to already known PID genes, assuming 5), and the most widely used approach involves selecting known the most highly ranked genes to be the most likely to cause PIDs. disease-causing genes as candidate genes for further investiga- However, a list of novel potential PID-causing gene candidates, tion. Thus, in this approach, variants are identified as candi- rigorously identified on the basis of biological relevance to the date PID-causing genes only if they concern genes already listed PID phenotype (both by HGC-predicted biological distance to among those known to cause PIDs (229 such genes have been PID-causing genes and by the relevance of the candidate gene’s identified to date) (6, 7). This approach is simple to imple- biological function to PID), would be useful and easy for most ment, but is likely to miss the true PID-causing gene in many PID investigators to use. As the HGC-predicted gene candidates PID patients, as novel PID genes (i.e., not already included may include some false positives, due to the prediction algorithm among the 229 genes known to cause PIDs) would potentially being based on protein–protein binding interactions rather than be ignored. Alternatively, the investigator would face the non- trivial task of inferring relevant PID-candidate genes from a list of hundreds following rigorous variant-level filtering, mostly by 1http://lab.rockefeller.edu/casanova/HGC

www.frontiersin.org April 2015 | Volume 6 | Article 142 | 1 Itan and Casanova Novel primary immunodeﬁciency candidate genes

biological functions associated with the gene, it is important to fil- known PID genes. For example, the gene CPN2 was predicted by ter gene lists further, according to biological relevance, at the final the HGC biological distance to PID core gene IKBKG to be a PID stage of in silico prediction. candidate. However, CPN2 has the biological function of protein We generated a list of in silico-predicted novel PID-causing stabilization, which is not a biological function that is enriched in gene candidates, which we describe here. We first determined the known PID genes; therefore, CPN2 could plausibly be removed biological clustering level of all PID genes, then used the HGC to from the initial list of PID-candidate genes. We therefore first used extract the top 1% of all human genes biologically closest to all DAVID (10) to estimate the gene ontology (GO) biological func- known PID genes. Finally, we selected as PID gene candidates only tional enrichment of all known 229 PID genes. We selected only those genes with relevant biological functions, similar to those of those functions for which p < 0.05 (a total of 462 GO terms). We known PID genes. We also show that the list of novel PID-causing then applied DAVID GO biological terms analysis (11) to the PID gene candidates generated contains 17 of 21 recently reported PID- gene candidates, and selected only genes associated with at least causing genes (not included in the database used for prediction 1 of the 462 PID GO terms. This resulted in a final list of 3,110 purposes in this study). in silico-predicted novel candidate PID genes.

MATERIALS AND METHODS PHYLOGENY OF KNOWN AND ESTIMATED NOVEL PID GENES BIOLOGICAL PROXIMITY OF PID GENES: STATISTICAL SIMULATIONS The biological-interrelatedness between the 229 known PID genes We determined whether the 229 known PID-causing genes (6) and the 3,110 candidate PID genes predicted in this study was were significantly closer to each other, biologically, than to other estimated with the functional genomics alignment (FGA) phy- genes, by estimating the biological distances between all these logeny (5). We first created a biological distance matrix between genes (a total of 22,366 biological distance values) with the HGC all known and predicted PID genes; we then applied a neighbor- and associated server (5,8) 2. We then randomly sampled one mil- joining algorithm, in the R statistical programing language APE lion biological distances from the set of all human genes (9), and (12) (analyses of phylogenetics and evolution) package nj func- determined, for both the PID and all-gene sets, the proportion of tion. Finally, we plotted a phylogenetic fan-shaped tree based on distances falling into the following categories: (1) small biological HGC-predicted biological distances between known and predicted distances (<10); (2) small-medium biological distances (between PID genes, using the plot function of the APE package (12). 10 and 20, 20 being the median biological distance between all human genes); (3) medium-large biological distances (between 20 GENERATION AND PLOTTING OF GENE NETWORKS and 30); (4) large-very large biological distances (between 30 and For visual depiction of the distribution of the 229 currently known 40); and (5) very large-extremely large biological distances (>40). PID genes within the full human genome, we used the predicted We then compared the biological density of PID genes with that HGC biological distance between all human genes, selecting only of other human genes, by first calculating the median distance direct biological connections (5, 13) between genes, to lower between all 229 known PID genes, and then simulating sets of 229 the complexity of the network. We then used the NetworkX randomly chosen genes, calculating the median distance between python package for complex network analyses and visualizations these genes, and estimating a p-value by determining the pro- (14), applying the spring layout function, which estimates the portion of simulated random gene sets with a median biological localization of each gene in a two-dimensional space with the distance smaller than that for the known PID genes. Fruchterman–Reingold force-directed algorithm (15). The nodes of known PID genes were plotted three times larger than those of INITIAL EXTRACTION OF NOVEL PID-CANDIDATE GENES the other human genes. We used, as an input, the connectomes (see text footnote 1) of the 229 known PID-causing genes (6), a gene-specific connec- COMPUTING RESOURCES AND PROGRAMING LANGUAGES tome being defined as the set of all human genes ranked according This project was performed on a Mac Pro machine with 12 cores to their biological proximity to the gene of interest. From these and 128 GB RAM. Biological distances between genes were cal- 229 gene-specific connectomes, we extracted only the genes in the culated by the HGC and server. Data extraction, statistical simu- top 1% of all human genes by p-value, to obtain an initial list of lations, and network analyses and visualizations were performed 29,885 genes. We then processed the novel candidate gene data in with the Python programing language. FGA phylogenetic analy- two steps, to prevent redundancy: (1) we removed all the predicted ses and visualizations were performed with the R programing genes already in the list of 229 known PID-causing genes; and (2) language for statistical computing3. The programs and online we retained only one instance per novel PID-candidate gene. These server used in this study (in particular for ranking candidate genes two filtering steps decreased the list to 5,012 non-redundant genes. by biological distance from core genes and FGA trees generation) are freely available to non-commercial users with step-by- FILTERING OF PID GENE CANDIDATES BY FUNCTIONAL ENRICHMENT step instruction at http://lab.rockefeller.edu/casanova/HGC and We then narrowed the list of genes down to those of biological http://hgc.rockefeller.edu, and the scripts for the minor technical relevance, by hypothesizing that a novel PID-causing gene would procedures in this study are all available from the authors upon be likely to have a biological function similar to those of already request.

2http://hgc.rockefeller.edu/ 3http://www.r-project.org

Frontiers in Immunology | Primary Immunodeﬁciencies April 2015 | Volume 6 | Article 142| 2 Itan and Casanova Novel primary immunodeﬁciency candidate genes

RESULTS biological distance between human genes,we found that PID genes DISTRIBUTION OF PID GENES WITHIN THE WHOLE HUMAN GENOME tend to be in the central hub of the human genome network, with We calculated and plotted the network of all 229 known PID- only a small minority in the central hub’s periphery and none in causing genes related to 50 PID syndromes (6) in the context of the the extreme periphery of the whole human genome. PID genes full human genome (Figure 1). Using the HGC-predicted direct are forming several tightly intra-related sub-clusters (i.e., in most

FIGURE 1 |The human genome and PID gene network. This ﬁgure shows all 229 known PID-causing genes (red) together with all 14,131 human protein-coding genes for which HGC-predicted biological distance information was available.

www.frontiersin.org April 2015 | Volume 6 | Article 142| 3 Itan and Casanova Novel primary immunodeﬁciency candidate genes

cases, a PID gene will have as a close functional neighbor at least which p < 0.01 for connection to the respective PID gene (29,885 one other PID gene) across a diversity of biological mechanisms, gene occurrences, many genes occur more than once due to their demonstrating the variety of genetic pathways underlying PIDs. close biological distance to more than one known PID gene). We then removed redundant genes, and genes from the set of 229 SMALL BIOLOGICAL DISTANCES BETWEEN PID GENES known PID genes. This left us with 5,012 non-redundant genes We tested the hypothesis that PID genes are functionally close to not previously identified as PID genes. each other (which would be the prerequisite for identifying candidate PID genes on the basis of biological proximity), by comparing FINAL SET OF PROPOSED NOVEL PID-CANDIDATE GENES the biological distances between PID genes and all human genes, We hypothesized that most of the novel PID genes would be likely in terms of the proportions of biological distances assigned to to have biological functions similar to those of known PID genes. five categories, from the smallest to the largest (Figure 2). Most Wetherefore used DAVID(10) to assess the functional enrichment, intra-PID gene distances belonged to the very small-to-medium by biological GO, of all known PID genes. We retained only GO categories (28.4 and 53.6%, respectively), the proportion of PID terms (11) for which p < 0.01 (Table S1 in Supplementary Mate- genes falling into these categories being larger than for all human rial). We then applied biological GO terms analysis to the 5,012 genes (7.7 and 40.0%, respectively). candidate genes identified as described above, and extracted only We found that the median biological distance between known those with a biological function already identified among known PID genes was 12.1, whereas that between simulated sets of 229 PID genes. This generated a final list of 3,110 in silico-predicted random genes was 20.4. None of the simulated sets of ran- novel candidate PID genes, identified on the basis of biological dom genes had a median smaller than that of known PID genes distance from known PID genes and a similar biological function, − (p < 10 7), consistent with the hypothesis of tight functional and described in terms of their relatedness to their biologically interrelatedness between PID genes. While it was expected that closest PID gene (Table S2 in Supplementary Material). We then genes belonging to the same pathway would display small bio- carried out hierarchical clustering of all known and predicted PID logical distances between each other, confirming this hypothesis genes (5, 12). This analysis showed that the candidate PID genes makes it possible to infer novel PID genes based on HGC-predicted identified in this study were evenly distributed over the whole biological distance to core PID genes. range of known PID genes (Figure 3), while they form together a network that is tightly itra-related biologically and functionally INITIAL ASSESSMENT OF NOVEL PID GENES (median biological distance of 11.0, compared to 20.4 between two Based on the demonstration that PID genes display close biolog- random human genes, p < 10−7). ical proximity, we hypothesized that currently unidentified PID genes would be located at a small biological distance from known ASSESSING THE PREDICTIVE POWER OF NOVEL PID-CANDIDATE PID genes. We therefore acquired the gene-specific connectomes GENES of all 229 known PID-associated genes and extracted only those for We retrospectively demonstrated the utility of this approach for identifying novel candidate PID genes, with 21 PID genes that were recently shown experimentally to cause PIDs (7, 16–38). None were included in the list of 229 known PID genes used in this study. Seventeen of these 21 genes were included in our list of proposed 3,110 novel PID-candidate genes, with p < 10−7 (estimated by random sampling computer simulations). Moreover, similar prediction rate of 17 of 21 is achieved with the extended list of 5,012 candidates, demonstrating the usefulness of the PID gene candidates short-listing procedure presented in this study. Table 1 describes these 17 genes in terms of their relatedness to the PID gene from the connectome of which they were identified.

DISCUSSION We describe here an extended list of 3,110 novel PID-candidate genes, initially predicted in silico with the HGC biological distance concept, then on the basis of the relevance of their biological function. Weshowed that PID genes (6) were signiﬁcantly closer to each other biologically than other human genes. We then extracted the

FIGURE 2 | Comparison of biological distances between PID genes and 1% of genes biologically closest to the known PID genes and fur- all human genes. Biological distances between PID genes (red) and all ther retained only those genes with a biological function similar human genes (gray), according to the proportion of distances in ﬁve to those known to display enrichment among PID genes. We gen- categories: (1) small biological distance (<10); (2) small-medium biological erated a ﬁnal in silico-predicted set of 3,110 human genes, which distance (between 10 and 20); (3) medium-large biological distance may be considered reliable candidate PID genes. In other words, (between 20 and 30); (4) large-very large biological distance (between 30 and 40); and (5) extremely large biological distance (>40). we predict that there will be a high proportion of PID-causing genes among these 3,110 genes. We then demonstrated that 17

Frontiers in Immunology | Primary Immunodeﬁciencies April 2015 | Volume 6 | Article 142| 4 Itan and Casanova Novel primary immunodeﬁciency candidate genes

MED13L MED10

MED9

MED19

HNF4G

RORB

MED27

EDARADDNR2F6

MED11

NR4A2 NR2C2

MED22

MED30

NR1H2 A

T MED18 MC

ENY23

N INTS3 POLR2J 7 PA

9FA

RARB B

ERCC1 MED8 POLR2K XAB2

POLR2I NR2E1 CTDP1 CTL6B

MED26 A NCR1 PHF12 ERCC5 T HDAC11 GTF2E1 IKZF2 T RNGTT IKZF3 GTF2H3 AF11 NR2F1 ADA ATAD2A

1 DPF1 G ERCC4 MED4 GTF2E2 T A KLF16 TXNL4A POLR2F F POLR2D IKZF1

5 CLP1 RARG SIN3B CDK19 L DPF3 SUDS3 ERCC8 C10

DDX23 A T GSC AF4 RORA NFKBIL1 SMARCD2 HDA

2F MED7 ERCC2 MGEA5

TFF3

T POLR3GPOLR2G TAF10 POLR3F BHLHA15 CBX8 SF3A3 CDK2AP1 ARID2 DAC8 SPI1 CHD7 FA MBD2 CAMTA2 MED20 RXRG SIN3A H RFXAP

SUPT7L ARID1B 5 RFX5 RFXANK CAMK1 THOC5 ERCC3 THOC1 HIST1H2BM E2F2 H2AFJ GTF3C4 MXD1 GEMIN4 RBM5 LANCL2 SAP30 SNUPN NR2C1 MBD3 ZFP57 SMN2 GTF2A1L MED13 CHD9 SMARCD1 CHD3 SUPT3H SMARCD3 SUV420H2 SIP1 TGS1 ASCL3 GLI2 EHMT2PRDM1 BCORL1 CBX1 SPOP DNMT3B ZNF830 TRERF1 NAP1L3MBD1 ARID4A CDK7 NAP1L4 E2F3 HIST1H4B POLR3A SETDB1 HIST2H3C PHPT1 SNRPG NFX1 MNAT1 PIK3C2A TBL1XR1 PPARD CAMK4 DNMT3L SNRPF NR5A1 NR6A1 KLF11

MED16 M LMX1A DLX2DLX5 CABIN1 RNF19A MIER1

EBF1 APG AC5 MECOM SUFU GLI3 SUV39H1 CIITA H3F3B PSMC3IP A4 D GTF2A1 DDX54 NAP1L2 RBL1 USP12 T ZNF354A RB1 H SNRPC POLR3C CHAF1A TRIM23 MPG NRBP1MLF1 HAND2 C4 WDR77 CYF DNAJB5 FUS CTNNBL1PTGES2 CAPN10 ESCO2 HELLS GAT MAPK7 H3F3A MED12 FA TBX1 RBM4 NCOA2 X2 OXC8 MEF2D ATRX SMARCC2 HMG20B CBX5 CHD1 NOTCH2B RCOR1PHF21A TA MYF6 MEF2A IK 7 MYF5 TCF21 MECP2 H MSX2 HNRNPUL1 GTF2F2 UD31 THRB MAML3 ADRB3 HDA

N PAX5 CLNS1A DA2A 1 SNRPD2PRPF6 PDIA2 X2 LEF1 POLR2L CCNH MEO MAFF TRIM33 MEF2C DNMT3A MAFG RBBP4 VIP 5 MED14 CARM1 LGALS7 TRIM24 PP SO NCO A6DAP3 NRIP1 ARID1A AC7 NF1 SRPK2 PTMS DBC1

PMVK S TBP CSRP3 ARP2 TRIM28DNMT1 ID1 MSX1 TBX3 MAML2 RXRA 2 ARA TFCP2 ELK3 A POU1F1 GTF2F1 NR1I2 P MITF MYOG NKX2 SENP1ELK1 HD ZBTB7A APBB1 HDAC9ZMIZ2 IVNS1ABP FB SAP18 CSNK2A2 PTPN18 FLI1 PPP1R8 PRDM4SERTAD1 PIAS4 OXH1 NCOR2 PAK6 SPEN SRF NKX2 CHRAC1 BY MVD F CYP1A1 PP 1PI NFKBIZ OXB3 A ER RBPJ UBE2W ESRRB ANAPC13 CITED2 NFE2 NHLH2 SMARCA2 KDM1A HOXB1 MED6 CITED1 ZHX1 PRO POLE3 1 CHEK2 SMARCA4 H VP 031P ZEB2

R NDN ARG FN CUL7 AX8 S ZHX2 MAML1 FZE HNF4A T PADI4 BN AC1 SNIP1 T S NR1H4 P 4 NCOA4 GCH HES1 ETS2 TAT2 NR2F2 AF12 GPS2 CIRBP AF6L P RBBP7 D TCF12 LCOR TLE1 X CHD4 CBFB SNRPD3 E TRIP12 TBX2 TGIF2 SF3A2 X1 A H MEIS2 PIAS1 N

H OXB7 MED24 HMGB2 L

F TCF3

P ESRRG DIRAS3 CREB5 YA CS T L SMARCE1SMARCB1 FHL2 H L CREB1 MDFI PBX1 NR2E3 FOSB Q CEBPE

MVK

KLHL13 ARNT2 C TAOK3 IRF9MAF AXIN2 S KCTD13 USP22 NMI HHEX KDM5B CRX NF ATA5 TEAD4 HEY1 TEAD3 SERTAD3 DR1 HOXD8HOXC6 SNW1 TWIST1 POLR2E CDK8 EP300 GATA6 G LDOC1

F2U

MED17 AIP DUSP22

ESRRA 2 RBBP8 PCBD2 RAB3B ZNF148

H CREBBP GABPA TA2 ESR1 TCF7L1 ATXN7 C2

PPP1R13L ZFYVE9 C1D GSK3A MAPRE2 E2F4 SP3 THRA M GTF2I TRRAP M TADA3 ING2 PDX1 POLR2C KLHL21 CBFA2T3 HOXD3 ZFPM2 LMO2 ONECUT1

G HOXD9 GATA2 MED21 GFI1 PARP1 NCOR1 HDA ARNT NRF1 DPF2 YAP1 RP 1A CTBP1 AHR

TNIK

CCNK A2 THOC4 TCF7L2

PURB 7RCHD B1I TNFAIP1 SMAD4 LGALS1 NRD1 KDM5A CEBPB MED23 RUNX1T1 AGER ELF3 EGR1

GRN RINT1 ZBTB16 CD7 SON MED1 F RARA

DLK1 SPAG5 SMARCC1

POU2AF1 SP1 NR4A1 VDR G TOP2B BAZ1B ATN1

RGP DDIT3 STAT4 GIYD1 STA TRIP6

DAC3 TF3 SND1 7D

2FRI RP NCOA1 CUL3 A R H BHLHE41 GATA3 T6 LDB1 MYOCD AGE BHLHE40 PML BTBD12A4 RB SMAD2

FNR KLF13 KLF1 CCAR1 HOXD11 CHRM1

4 CSNK2A1P

PURA 1 KLHL12 KDM4A FGF6 KAT2A

L1L CTNNB1 MEAF6 NCOA3 K PHF17

S CHRM5 MED15 NLK GATA1

LK L TAL1 FGF1AES TAT3

F SF3A1 TCH1

4 RAD1 ESR2 6F

DMC1 HOXD10 O RAD52 VL2 SLC26A9ING5 RPA3 ATXN1 KAL1 KAT2B DVL3 N DULLARDPPP2R3B MLLT1 D AFF1 POU2F1SREBF1 POLR2H CHRM3LRRC29

MYST4 RAD9A VANGL2DOT1L PP MYST2ING4TP53INP1 CCNT1 TACC3 BRD4RAD51XPA HSPB2 MY IFI16

SHOC2 CRYPOLR2B TCF4 IFT88 RECQL5 ARGC1A EZH1E2F6 HMGB1FGFR1 OD1 ID3 2E RELA RAB26 ACP5 COBRA1 ABSNRNP200 BANP ULBP1ULBP3 PPP1R15B PPP3CB 1F CELA3B ELP3 SUPT16H PPM1L EED RASGRP3 RPH3AL ROCR CTTNBP2 PHB2 P SSRP1HSPB1 NR3C1 RAB3C JMY PHB CATSPER1CATSPER3 PPP3R1 RWD1 DYNLL1 ADAR B RB1CC1RNF111CDC73 CBX4 RAET1E PPM1K NFIB PRPF19 TP53BP2 RAB3D GTF2H1 SUZ12 HRK PPP2R3A CDK9 BMF PFDN6 NFIX DHX9 PRMT1G3BP1 BAD CABP4 ZBTB24 RUNX1 RAET1G CA ZNF346 TP53 PARP3 FKBP1B XPO5 ETS1 PCGF2CTCF CNA1F RPA1 CSNK1A1 EZH2 CA2 ILKAP BCL2L2 RAET1L RFX1 CUL9 UNC13B PPP3CA ILF3 SF3B3YBX1 PPP2CB UNC13DMRPS26 TIAF1 RING1 STX11STXBP2 ULBP2 POLR2A APITD1 PDP1 PPP2R5C SKI TRIB3 SLC18A2 FANCF NAE1 YAF2 PMS2L1 SMARCAD1 CA KDM2B PMS2L5 PPP1R13B RAD51L3 CNA1G TLX1 STK24 NFIC CD1A PPM1E PDCD10 SKIL STX1B CPLX2 FANCE PBRM1 ADRA1B S100A1 XRCC2 PMS2L2 PPP1R9B TOR1A RFWD2MDM4 BCL2 RIF1 FAP1L2REM2 C FPRDX6 RMI1ANCL MR1 A ACNB4 AM175ASMAD3SF1 F ZNF496HP1BP3SP100 PPM1F BRCC3 LILRB3 PPP1R2P3 S100B DAXXHIPK2 SNAP25 OXA3 POLR3DMX1PPM1J NEDD8G3BP2 BIDBAK1 XRCC3OP3A PMS1 H MADDUBA5 MDM2 BAX VAPB AMP3 T ANCB C PPP2R2A NDRG1 V F F IPO13 ACNB3 PPP2R2B S100A6 STX4 CLK3 ILF2 SENP2 UIMC1 BMI1 STX1A FANCG PMS2 HLA DCLRE1C STK3 TP73TP63 CA NPM1CDKN2A YBP T1 TRIM16 TFDP1YY1RATF6 SLC6A1 CD1B PSME2 HIST2H4BHIST1H4K TDP1PNKP CNA1E CLK2 PO PSME1PSMG2PSMG1PSMB10PSMB11 LIG3 UBE2I MST1 GCH1 MSH3 TOPORS PPM1D APEHDDX5 RNF2 RT TPP1 PSMF1 TDG HOMER1 ANXA5SHMT1 ATP1A2BCL2L1 VAMP8 ANXA8L2 SUMO2SATB1 PRKAB1 BRE TXLNA FANCM HIST1H4J POLM BLMHATP2A1 THEGPRKAA1RBP3 MLH3 FCG ACD PSMA8 LIG4 RYR2 PPP1CA SPHK1 BLM XRCC6BP1 PPM1B USP7 RT7 HIST1H4D SETMAR KRT14KRUBA2 STRN SAPS3 STEAP3MCL1 SAE1 USP16KPNA2CSE1L SISIRT6ATG12TG7 RAB37 GAZGP1P1 PSMD5 NHEJ1APTX T5 TFAP2ATPT1 A MLH1 E SHFM1 ATP1A1 PSMD13 HIST1H4H NAA16SP110 MID1 T2 RIMS1RAB3ADNAJC5 STAC PRF1 TERF2IP PSMB6 TNFRSF17 PPM1A SIK2 IRF3BCL2L11PIN1 STXBP4 ZBTB32 HLA TNFSF13BCLUL1 SUMO3SUMO1P3 CYCS RTG5 BRIP1 HLA PSMB9 CACNG2 SI A FANCCFANCA TAPBP POMP DFFBDFF MARK4MARK3 OBPKLRK1 SYCP2 ATMIN HIST1H4E RAB34 PPP1CC ING1 PLK3 STAG3 NUSAP1 PSMD10 TLK2TLK1 TNFRSF13C EGFL8NCR3 PPP2CACCND1IGBP1PIM1 CDC25C STXBP3 MBD4 PSMB4 A CCNG1 TYRTAP1 SYCP3 SGOL2 PSMB7 HIST1H4C CHD1L PPP2R4PDP2 ID2 RASL10A KPNA3RAG1 APLF CACNA1C TCP1 DDX20NR1H3 SIRT3RT1 MSH5 PSMB3 ASF1A NUP62KPNA1 SMN1NR0B2EEF1D SYT1SYT7 MSH4 AZGP1 PSMB5 BANF1 CD59NGDN PPP1CB HUWE1PPP6C MAP3K15SI KIR2DS3TAP2 SMC3 USP4 DCLRE1B HIST1H4FNFKBIL2 TNFSF13 PPM1G RBL2 MTMR15 KIR3DS1KIR2DS1CD160KIR3DL2 RAD50 CECR2 RASSF1 UBE2T TNKSTNKS2 ATM ASF1BHIRAHIST1H3E MST1R MYH6 B2M PSMA3 RSF1 CHAF1B RANBP2 TUBB OXO3 ABCF1 CALR TPR VPS72 SFN FOXO4F RAD21 MRE11ARRM2B HIST1H4L BRCA2MSH6USP11MSH2 FANCD2LNPEP PSMB1 NUP85TPX2 ANP32AXRCC4 CPEB1SMCHD1 C8G BAZ2A TXNIPKIAA1967 SYCP1 C8A HIST1H3B OXO1PA8 HIST1H4I HS F GPR77 APOC1SAA2 SMARCA5 HIST1H2AE HS PSMD12 AIRE NUP98KPNB1 POLL EDF1 PA2 HERC2TXNRD1 SLC2A4 SMC1A BAT1 RNF8TSC2UBE3A HFE NBN PSME3PSMD9 C5 H2AFY PSMD11 CETP SET ARL2 AQP2 HIST2H2AA4 XRCC5 TOP1 TINF2 TERF2 C8B EIF2C1EIF2C2 NONOSFPQ PSMB2 APOL1 ERLEC1 XPO1 C9 HIST4H4 DNAJA2 USP34 NME1 WRN PSMC4 NCAPD2 NASP XRCC6DNTT DCLRE1A TERF1 LC SMC2HIST2H2AA3SERPINF2 AAAS ZIC2 C7 AT HIST1H4AHIST1H3A PSMD8 NR3C2 NUP205 NCAPH DDX1 C6 JMJD7 MPO APOC2APOA4 HIST2H4A CHEK1 PSMC6 C5AR1 CAPZA2USP33 NEUROD1 PLA2G4B ADD2OS9 CBX3 PSMD1 KLK4 APOA1AP NCAPD3 IPO5 ENO1 O EGLN1HIST2H2AC RAN EIF5A PSMD3 PSMA5 KLK5 A2 MTPN HPR APOC3 NEURL NCAPGHAT1 CPDKIAA0368 PSMA1 ZC3H12A PDFCOG4NAPG PSMD7 ERO1L CUL5CAPZA1RACGAP1P TXNL1 PSMA6 HP ERP44DNAJC10 HIF1ARNF168 SCFD1 PSMD14PSMC1FKBP8 CST7 MMP25SOCS6 SMC4 HIST3H3 PSMA2 LCN2 SH3BGR STMN3ZER1CUL2CAPZBHIST2H2BE PSMD6PSMC3UBE3C PSMA4 SLIT3 SUMO4HSP HIST2H3A NSFVPS45 PLP2 TMPRSS11A CTSC TCEB1 H2AFX PSMA7 UFSP2 CASKID4 LIPCOLR1APOE A5TCEB2 E PSMD4 SEMA7A STK38MLL5 PTGDS ANGPTL4 ERN1 P PSMC2 ITGA1 MSL3 NFRKBINO80 MXD3MNT CAPN3 APBA1STXBP1APBA2 PDIA3 AS1 AP3D1 PSMC5 LAMA2 MYST1SSH2 TJP2 LPL CLU VHL AMP7VTI1B COL3A1 RUVBL1MXD4 TINAGL1 SYVN1 V HMOX1UFD1L LAMC1 COL1A1SPARCACHE SMARCA1CHD8HCFC1 ARPC1BARPC5 LMAN1L COL1A2ITGA11 SPON2VNN2 PRPF31 MXI1 TF ASB2 SQSTM1MAP1LC3APDE4A ITGA2 PROC BRD8EPC1 BPTFSETD1AASH2LCUX1TAF9 TOP2A KEAP1PSMD2DDB2XPC FPR1FAP YE ACTR3 PSEN2VLDLR P4HBFEM1B USP14 CD9CLDN1TIA1 FIGF F10 STMN2RGS6ATS4 USF1SETD7TAF6 DYSF CXCL10CXCR3 COL2A1 VEGFC FPR3 FGGITGAD ITGAM MMP9TIMP2 EP400 ACTL6A WASF3 S100A11 RAD23B CD36 FLT4 FPR2 FGAFGBF13A1 ICAM5 ITGA8 DMAP1PLA2G4A CAPNS1SRI ANXA1KRT18KLDLR TMC8BNIP1 PF4 VMDKASPTUBB2BXPO6THBS1 PVR ITGAL CAPN1RT8HSP90B1 HIST1H1C NTRK2 CCR5LACRTCCL5ITGA6 ITGB1 PL AURAU ITGB6CD226ITGB2 STK35CCND2 SRCAP ARPC3 ZFAND5 SDC1CXCR2 VTNANGPT2PL THY1 CCND3CDKN2D MYC TUBA1BALDOAHD A C6 SLC30A1 CXCL12ITGA4 KLK13 CYR61 CDK6 MYCN A CTR2 ATXN3DNAJB2CD164IL8 ADMAMBPADAMTS1A2M AV DAMTS4 BCCIP MLL SOD1CCS ITG AFCN1ELNDAMTS3ASCGB1A1MT2AMIA CDKN3CLK1 KAT5 T TARDBP APOBEC3G RAD23ACXCR4SDC4 PLAT LOX ANPEP CDC7 ING3 AF1 ITGB3 SERPINE1CRP DBF4 RUVBL2 SENP3 PLGIGFBP3IGFBP5 GAP43 PLA2G2A ORC6L SMARCAL1 CDK4 MORF4L1 UBE2ODDI2 SPP1 TGFBIDPTAHSG ORC5L GADD45BGADD45G CCNA2 MORF4L2T DRA ORC4L ATRIP CTCFLHIST1H1A AB1 CD63 FN1 CTGF GINS4GINS1 ORC3L CINP MAP3K4CDKN1BSPCDK2 USP18CKS1BSSBP1SKP2IGF2BP1DSTNACTG1ACTB SDCBPHLA CDC45 ATR GADD45ACCNE2DYA DRB3LOC100294189LOC100294278 MCM8ORC2L HUS1 CCNE1 LOC100294277HLA GINS2MCM5TOPBP1 CDK1CDKN1A KDRMYOF CD70 LOC100294276MS4A1DRB4 RAD9B MCM6 RFC2 PFN1 CD27TRAC HLASPG21CD8BP T1 RAD17 MCM4MCM2 CDC6 TGM2GSTP1 CD2 LNX1CD5LNX2 PLSCR1CD3ECD3GSLA2 TRACD8B CHTF8DSCC1TIPIN MCM3 ORC1L CD40 ZAP70 RFC3RFC5 RNASEH2CRNASEH2B ADRBK2COL11A1 CD209CEL CD8A CD1DPTPRCPTPRCAP XRCC1NEIL1 RNASEH2A MCM7 SERPINH1CD3DCD247 SKAP1 ALKBH2PIF1 POLK RFC4 RFC1 MCM10 HIST1H1B LCK REN CCNO DDX11FEN1 PCNA LIG1 CDT1LIMK1CFL1 TREX1 MAPK15 CDKN1C CAMK2BCAMK2DCAMK2G RALARALB GML STK19 OTC MUTYHAPEX2 CD4COPA CALM2 FGF13RIT2 CAMKK1PLA2G6RAB40AL POLE2 POLE TXNREV3L POLA1KDM5C CALM3MYO1EAOCALM1X1 PRIM1POLA2PRIM2 UBR5PAIP2PNPTFTXN2AMUCHL5ADRM1NOS2PRDX1CTSD APRT PPEF1 AAGTCE2 RAD54BPOLD2DNA2 SMG1 RAB5CHTATIP2RIN2RAB5A APPL2RAB22A POLD1 POLD3POLD4 RABGEF1RAB7A RABEP1KIFAP3 HACE1RAB4A APEX1REV1POLI POLH HLTF OPRM1DAD1MAGT1PLD1AP3B1ARF6 RAB1AGULP1ACAP1 TUBG1POLB CDC16 AURKA RAD18 UNG GCKRRRM1PMPCAUCHL1TUBA1AFLT3UQCRC2RTN4 RAB1BARF1RAB11AGDI2RAB2A ADAM9MAD2L2MAD2L1CUEDC2 CDC20CDC27 TTK IDO1 ZRANB1RPL24USP44MTHFD1 LRBAREG3AHERPUD1SNCASNCAIPSIAH1APCTUBA4ACACYBPSIAH2ZIM2 NPC1 TRHDEREG1ACLPTM1L SERPINA1 VCPIP1 GOLT1BTLR8 CCDC99AHCTF1 NUDC INTS1NFE2L2PMF1 HIST1H2BB AUP1UBE2J1 CLPTM1UNC93B1 TAOK1NDC80MIS12 ATF4 SORD UBXN1FAF1 UBXN6ANKHD1 EIF4EBP3 TTBK2 CENPHCDCA8NDEL1KIF18A BUB1CENPA ZWINT EPHA1UBA1MAO FOS STRBP HMGCRERLIN2NPLOC4UBE4B CAPN6 ABCB1CDC123AKLF5HSPH1ZBTB7B TTC7A AMFRUBE2G2INSIG1RNF5UBE2E3YOD1 VCP UBXN4 KNTC1 ZW10TBL1XPAFAH1B1 BUB1BAURKBUBE2E1 PTTG1CSF3R ARIH2 GTF2BRIPK3REST FOSL1 SLC46A1 HMGCS1JMJD6 DERL2 KIF2C BIRC5CENPE BUB3 DNAJA1SETD8PLAAFZR1 USP5USP25 PPIAATP5A1BSG CNTN2OPTNPYGL JUNB JUN VPS13B KCNJ11DERL1 TUBG2SS18SYNJ2 GTSE1 ARL3CLASP1 CENPKSOX6 UBE2SUBE2COTUD5GMNNUBA52UBBUBE2KUSP28USP21 MAPKAPK3USP13BCAP31FIS1 TUBB4CDC37CDK11BSTK11ETV4 RTEL1UBC CLN6 ROBO1SLIT2 ABL2CLASP2 ANAPC5ANAPC1ANAPC7 TRIM5KIR2DL4 TNNT2SERPINC1NOS1BCL6BCOR JUNDFOSL2 HBZ TMC6 ANAPC11ANAPC10 ANAPC2 UBE2HRNF216 PEBP1IREB2TRIM63TNNC1TRIM54OTUB2ATP2A2 FOXP3N SLC37A4 CDC26CDC23ANAPC4 TRIM32 ATP5BCFD NFAFOFOTC2FXP1AT5XP2 RCAN1IFIH1PIAS3TRIM27ABI2 PIKFYVE NFATC1 JAG1DLL1NAZI2PTCRAOTCH3DDX58 FIG4 IRF4 MIB1DAPK1 HAX1 NFATC3 ODC1RBCK1 NFATC2IP EGR4 RAB12 PLN TIRAP ABCB4DGKD SCARB1 SHARPIN HSP90AB1SLC25A4SAA1CPT1A IRAK1PINK1DUSP6PTPRZ1MAP1LC3BNOD2 USP19 NLRP12 MGMTEEF1A2HSPD1TLR4 CLN3 TRAF6 RET FRS2 EIF4EEIF4EBP1 ATG16L1GALNT2 TEK SHC1 T1 CBL SMAP1 IRAK4 GRB14 FL RPS3EIF3BEIF3E TICAM1 NUMBL ANGPT1DOK2 VEGFAXBP1 CASP3CD14STAR PTHLHEPO CDC34CREM CAPN2HSPE1NMT1CC2D1A TNFRSF11AEDA2R PDGFRB TOLLIP IL1R1 CAV2 NOS3 CAV1 CRKL NOSTRINCAV3 DNM2CNR1PDGFB PDGFRA V1 RNF11UBE2D3 IRAK1BP1IL1RAP PRCP TCN2 KSR1 VA FBP1 PDCD4 RAPGEF1PTPN2 ITK PLCG1 RPS14 IL1B PIK3R1 UBE2D1UBE2D2OTUB1 TLR3TLR9 KHDRBS1PTK6 CD86 FGD1 RPS15 TRIM69 IL1A RBM9 CD28 CBLBSH3KBP1 AIMP2 TLR7 CCDC47 HCLS1QKI CTLA4 PTK2 MET TRIM21IRF8 STUB1 CNPY3 MYD88IRAK3 SYKCD79B INPPL1 EIF4A3 IL1R2 CD80 TGFB1I1 SOS1 RPS5 SIGIRRIRAK2SARM1 ICOS DDR1 DCUN1D1UBE2N RPL11 LY96CABC1NLRX1IL1RL2 FCER1GCD79A AK2 SPTBN1 TLN1 BTRCGSK3B RPS19 TLR1SFTPD ICOSLG FCER1AIGHE VCL CDC25A TAZ RPS13 LIME1 LST1 NCAM1PTPRA IGF1R COPS3PTGS2 CUL1 UBE2B RPS6EIF5B TLR2 G2E3 PSLC6A3 ARK2 CCL22 SPTBN2 BECN1SH3GLB1 EPORKIT IL10RA RPS9SRP72 FASTKD5 SPTBN4SORBS1CALD1 TPM4 EGFR KLF10 S CSTBTBL3 PDLIM7FBXW11 RPL22 DCX MYO9AMYH10 RIOK3 TAT1 FBXO7 RNF20 PES1GTPBP4 TLR6RNASE2 ANK3 TPM1 TFRCTF IGF2 UBE2L3 NFASC F8 CAMPARF4 LRP6 GRIP1 RA RPS17 PROS1IGF2R INSGRB10 USP9X PWP2BMS1G2 RBX1 EIF6 TNNI3MYBPC3 CREG1 TGFA IRF1 DDB1 SKP1 UBE2V1 TNNI3K IDE IGKC VPRBP CNBP RPS24 IGF1NPPB SAMHD1 ZNF593 IGFBP7 GCG ITCHARIH1 RPS15A BRAP PICK1GOPC DTL TIMM50TNFRSF10B CS CRBN IL10SIRPG RNF40 SEC61B F3 GCGRIGH@ IL32CTSB SRP14 SGK2 TACST3 NOP2 DDOST IGHA2 CGB8 RIPK1 TRPC4AP IL10RBUBE2A IGHG3IGHG2PCSK9CGA TFPI IGHG1 IL22 SHPRH WDR12 UPF1 FSHBF12F7 TSHR C4B CAMLGTNFRSF13B SRP54 CALCR CFLAR CD48 FBXW2 IL22RA2 PAPD5 DEPDC1SSR1 TNK2AXL CUL4A FBXW7 SBDS EIF4G1 ISG15UBE2E2 FBXO4FBXO30 ETF1 SMG5 SRP19SRP68 PRKCBBCR NTRK1SCNN1G CAP1 CSF2RBAT RHOA MAP3K14 UB FBXW8 NIP7 NCBP1 PRKCSH 1 IL34FCGR3B SOCS1GHR L CSF1 TNFRSF10A PDCD5 EIF4B UHRF2CSF1RSPG20EPS15 AVPR2 OX5 IGFBP2 CUL4B FBXW5 GSPT1 IGLC1FCGR3AAPCS TYK2 VAV2 SPRY2 FASLG PELONCBP2MIF4GDFBXO18 IGKV4 IGLC3HPN MAP4K1C3AR1SLK EIF4E2 C2 CRHR2 NOL8 IL17A HGFAC PTPN11 EGFRY1 PTENNEDD4LSCNN1BKCNQ1USP2 IKBKBBCL3 IL18 TSPO 23 ST14 IGHG4 IFNAR1 ADRB2 FBXO2IL28RAIL29 LTMAGI3 IGLC2 IL2RB SP DUSP1MUSK L SMG7ECM1SKIV2L IGHM PTPN1LTK PRLR MAPK1TOB1 O3 TBR UBE2L6 CAND1 ALTB CDX1 IGHV3SPINT1 IRS1 RIPK4 ITM2B TNIP2 MASP2C4A UCN3 70 PRKD2 PTPN3TRPV6 IL7R PRMT5 CD74 TIRF7 ANK NFKB2 TNFRSF1A RBM8A DNAJC16 IGHD C1QA IL2RG TYR CASP9 RELB FADD FBXL15 ARR3TAFR NDFIP2NEDD4 TNFSF14 IGHV2 IGLL1 IRS2 MAP2K2 P ERBB4RNF41 CASP8AP2 CCL3 INSR DAB2 HGS MAEA SMAD1WWP1 HERC5 VPREB1 IFNW1 PRL APP TSG101 HIVEP3 RIPK2 WDTC1 LHX9DGKG HBEGF AKT1VPS4BOM1L1 TNFRSF1BCYLD NFKBIA UBA7 CASC3 IFNA13 IFNA2 BTC AP2A1 T MAPK3 MAP3K2 IKBKETRAIP TNFRSF25TRADD FBXL5 C1QB JUB APLP2 PRDX3 DDIT4 RUNX3 TAX1BP1 WIBG EREG MAP3K1 TRAF3 TUBA3CMAP3K8 B TNFRSF6B IFNB1 JAK2 GRAPGPS1 ARRB1 USP8 BIRC3NLRC4 DET1DDA1 C1QC ADRBK1 NHP2ERBB3 CFTRANXA2 EIF2AK2 HIPK3AG4 NF2MAP3K11SH3RF1 SNCG BAAIFM1UBASH3A HTRA2 IL18BP FBXO31 C1S IFNA5 BMP1 PRSS2 FSHR MAP2K7 R NOL3 TNFSF15 C1R NTSR1 SLC1A2 NCDN RHO MEN1UNX2 MAP3K7MA TRAF1 NFKB1 LRDD LMO4 SNCB TAB2 REL APLP1 MAP2K4 DNAJB6 G1 TNFAIP3TNIP1TBKBP1 IFNA6 IL6ST LEP APBB3 FZD1 EDNRB GED1 XRN2SCG2CASP8SMPD2 T5BT5A CCKLEPR MASP1RPS6KA5AP2M1PDCD6 MAP3K5 GRK5GTR1 TRAF2 AGAP2NFKBIE FASPTPRM IL18R1IL18RAP FBXO27 IFNA8 APPL1SGSM3LYST A F2RL1DUT DIABLO SMPD1 TA AKT2PTPN23 FTH1FTL SSTA BIN1 PIAS2HSPHS TRAF4NKRFDHX58 AK1 DPP3 FBXL2 MAPK8 ITGA5P B TGFB1 ANXA4 NGFR TRAF5TRAF3IP2 FAIM2 CASP2 J AP2A2 AP1 PA2G4 FGFR2 TNF XIAP TBK1ZMYND11TARBP2 CRADD IFNA16IFNA14 MBL2 ARK7A DOM3ZNFKBIB FBXO6 LIF IL9R PEX5 SAG F2R PRKCZ APBA3 EDNRAGNAL PLCB2 PA1ACE1 SMURF2 SI IL17RA PRND IL6R LRP A9 A1B BIRC2 ASB3 SOCS2CISH EDEM1 P JPH2JPH3 V CD40LGOCA2 RAB27B EBI3IL6 A3 SPG7 STRADB A1 NOD1 SOCS3 AP2S1 EFS GNG2 TRPV4 AR ABCC1 TNFRSF12A FBXO44 PCBP4 COPS2 TGFBR1 IL27RA GAS7 PLK1 L1CAM IKBKG OSM MAP1S GNA12 HS NR0B1 PRKG1ALB ELANE DNAJSH2B2 AFG3L2 ATP6V1H GLRX MAPK8IP3 ABL1 C1QBPLAMB1LLGL1 PRNP HBB CARD11 IL17F IFNGR1 YWHAE RANBP9 P RNF128 SLC9A6RASA3 AP2B1 MAP3K6 AXIN1 S100A9 IFNG MPL TNFRSF21 MAPK9 SMAD7ARD6A S100A8 HIP1 IFNGR2 ADCY8 ARHGEF1 COL18A1 CASP1 FBXL3USP47 HES5 SH2B1 PRSS3 TERT VIM SMURF1 IFT57 SH2B3 TSC22D4 FMOD CASP4 A USP31 FLNA LCP1 TGFBR2 BIRC7 IER3PGF ITPR1 ERG CHUK CARD8CARD6 LDLRAP1 CYP1B1 TCF19EDN1 HSPA4 CASR CVR1B SLC9A3R1SLC9A3R2 MALT1 CASP14 CASP10 ATP6V0A1 HRAS GJ OCLN SGK3 AATFXAF1 IL12B ATP6V1B1 ADCY2 WDR48 DCTN1 TSC1 F11R PAG1 VEGFB PPP2R1A RDX DEDD DUSP16 SGK1 HCK DEDD2 GNA15GNB1 A1 KRT1 SMAD6 YWHAZ F2RL2 A TIFA IL12RB2 FCN3 NEK2 IQGAP1B FCN2 MYO6 PPP5C EPHA5 CDH1 HSF1 CCNB1 CTN4 NUMB JUP CARD16 ARHGEF7 NGF ECE1 GP1BA AG3 IL12A YN RALBP1 IL12RB1SPHK2 HSP90AA2 HTR2A ARAF LRRK2 BCL10ERC1 YWHABZFP36BRAF G L PDZK1 BDNF TNFSF10 IL23AIL23R HSD17B10 RGS4 APAF1 CSNK2B PRKCA PARD3 SLC9A3SHANK2 GNAQ T19 PDRD2 TM2D1 FGF2 SNX6 GP5 RAB14 MSN KLK3 SSTR2 AWR SERPINF1 FGFR3CDH2 TOR SMAD5 TNFRSF18 RPS6KB1 CD46 GNAI2 IP6K2 KR CPB2 RAB10 SERPINB9 MAPK8IP1 FKBP4 A PRKAR1AICAM1 ACTC1 APBB2 YA2 CD93 GIPC1 MAP3K12 BAG5 PLK4 TNPO1 RIC YWHAP ITPR3 SORT1 MAP3K13 E GNAI1 MKNK2 RAF1 MTNR1A DSP M NGFRAP1SLC2A1 MAPK8IP2 RGS1 TSC22D3 SHH PRSS29P PEA15 AG9 CDC25B FGR ALPL MAGI1 MAP2K1 VA GP9 CD2AP ACF1NDUFB4NDUFV2 PTGES3 RAB13 PRKCG TGFB2 PYDC1 TNFRSF8 RPS6KA1SP FCGR1A PRKCI SUGT1GLMN PRKRIR CD44 BMPR1B YRP1 MAPK14 RAB6A GP1BB TJP1EPB41 UBE2R2 FKBP1AAIPL1 YWHAQ F2RL3 RAB8B F11 TNFRSF10DTNFRSF10C

STK4 APC2 FSCN1 MARK2 NRAS KLKB1 PANX1 PGL SOX7 FCARIGHA1 CASP5 RPS6KA3 MMP2 CD24L4 TRPC6 PSEN1 NLRP3 MTOR CDH5 PRKCD NLRP2 MAPKAPK2 RPTOR A PYCARDMEFV ZBTB33 ACVRL1 MAPK12CTNNA2PTPN7 CNKSR1 ZFP36L1EXOSC9 RAB8A AD ENG CARD10 NOP10 CTNND2 DPP4 TGFB3CVR1 DUSP9 THBD ADORA2B CD81 CBY1 BOC KNG1 MKNK1 PCSK2 PARD6B WARS KLK2 HIST1H2BC KRAS MAP4K2 PAK2 SERPINA5 SMAD9 MAP4K5 FKBPLDCTN2 A ATP F9 PKD2 ARHGDIA TNFRSF9

MLST8 RHEB TRPC3 C GH1 PRKAR2B TNFSF9 NCSTN PRKACG

RAP1A 3 ARFIP1 MAPK11 SIK3 MMP14 CARD9 AKAP8 AF1 A MIB2

YRK1A RRAS2 K AMH TUBA3D APH1A CVR2A ADORA2A PRKCE ARFGEF2 NAMPT DUSP10D ARHGAP4 F2 3PAM IFITM1 CLOCK ABY GAS1 CSNK1E A SELESELPLG GPC3 BDKRB2 AKT1S1 GRM1NEK9 CDC42 TNFRSF4TNFSF4 SIK1 C B SPN YWHAH RALGDS A RAB33A TA1 AT5 CENPJ ARHGAP1 PRPF4B FYN CVR2B CSNK1D TMSB4X MMP1 PRKCQBDKRB1 TIMP1 BMP4 DEPDC6 NEK8 ULK1 TRPC1 PRKCH ACVR1C ULK2 CD19

M NTN3 CORO1A PRKD3 1C FCER2 IGSF8 GNMT Y MAPRE1 RAC2

N R RAB19 O9B RASL10B RHEBL1 AR TUBGCP4 SERPIND1 BMP2 MSTN NCK2 OCK2 BLOC1S2 O SELL TNPO2 CDK5 BMPR1AGREM1 NDE1 2M ORAI1 CE EPHB2 AMHR2 FKBP1C X TSR1 ON TRIB1 CD34

IHHDHH R

1 IFNA21 BMP7 X

CC TSPAN4 1 FST

WHSC1 N 4 RHOUARFIP2 BMPR2 INHBA

C STIM1 CD300LG DOCK1 RGMA DLG1 SELP RGMB

F EPHB1

SRC PAK1 2

GRIN2B PKN1 FGFR1OP PAK3 AM ACTN1 PKN2 ARHGAP6 BAIAP2 GDF5 PDPK1 GRB2 A CFH

C SCRIB LPXN LBP CEP63 DAM10 DAM12 R LINGO1 O

3C A A STIM2 ABI1 ABR

BY T 1PB TUBGCP2 RAC3 IRK C3 EFNB1 CDK5R1 CR2 TIAM1 B R DTNBP1

3 RGS3

3 GAS2ESPN DCC ARHGAP29 DGH DLC1

K LMTK2 SPTAN1 CD55

IRF6 C4B K NCF4 1 I VCAM1 2 SERPING1 NOG PTPN13 IP CER1

CIB1 B BCAR1 4 P BMP6 GAB1 BMP15

GTI GRLF1 DAB1 GI2 A CD97 GDF6 DLG4 P CADM1 LRP8

PAK4 G1 ROCK1 GQI ARL1 A ACTN2 APOH ARF5 KCNA5 RHOQ GNG12 A MA DNMBP DPYSL5 CFP BMX RHOD NCKAP1 SEMA3A SOS2 WAS KCNA2

FERMT3 DMAP2 L2FC HRA

HCSIN CFI MAS1 PRKD1 WASL ARHGEF2 GDF2 ELMO1 AMPH ITSN1

DOCK2 RELN IFNA17

KDM6B

FARP2 NCF1 IFNA1 PLXNA3 GRIN2A

GRIK2 MYO1F M RHOBTB2 C4BPB

2SLA

SH3PXD2B RND1 1 PLEKHB1 YES1 FER FES STAMBP TEC PXN ADAM8

E RASA1 GSN C CFHR5 PLXNB2 ABLIM3 DOK1 AM15 ABLIM2 WIPF1 NRP1 S SHB NRP2 DLG3

EPS8 C LSP1

2 SSH1 A NPHP1 CD AD 4

MAP4K4 NTN1 GAA ARHGAP5 G NCK1 AIF1SH OI CD CDC JAK3 EXOC4 NCF1C PTK2B CRK PAT1 PLXNB1 UNC5C

GPB24C IL2RA V3 NOX3 CFB

R NIPA1 RVA FYB LCP2 DSCAM R AALK DOK3 CSF2RA CBLC T E24 LAT2 IL5RA PI4KB V OOM2 SHC3 PACSIN1 CD47 PTPN12 PA 1PE24 G GAB2 EPHA3 A P

ILK TRBC1 UNC119 2E CSF2 R 4 PIK3R3 SLAMF1 FOXN1 ID CFHR3

BLK UNC5B RAB9A1 S24 TAPBPL

DC GH ARHGEF12 SHC2 UNC5A BTK MTSS1 SEMA3C LY9 IL2 PTPRD GRAP2 PDCD1 D CSK PACSIN2 DOCK8 RHO PIK3CA EPHA2 PACSIN3 RHOJ PIK3CB EPHA4 I MERTK C SEMA4D DAPP1 RND2 G TREM2 CFHR1VSIG4

D SIT1

ARHGEF11 RHOG C

V BLNK DBNL PAK7 PTPN22 SEMA3E EFNA5 R PIK3CG EFNA1 EPHB3 NGEF 5 ELMO2 5 H

P 2 MAP4K3 O

BTLA K E24CD

IL3 H SH3BP5 SH2D2A ARHGEF6 IL4 T PIK3CD EPHA7EPHA6 S CMTM3T1 FCGR2B CD84 R ACP1 EPHB6 TNFRSF14 H CD38 PTPN6 LAIR1 IBTK PI4KA C IL5 IL13RA1 FO B4GAL SH2D1A R TLR5

BOH

SLAMF7 ATPIF1 B

HES3 1

IL13 GH IL15 R

IL15RA BO IL7

IL21R

IL9

IL21

FIGURE 3 | Functional genomic alignment phylogeny of known PID genes and novel PID-candidate genes. A phylogenetic tree of biological distances generated by the functional genomic alignment method, showing hierarchical clustering of all known (red) and predicted (blue) PID genes.

www.frontiersin.org April 2015 | Volume 6 | Article 142| 5 Itan and Casanova Novel primary immunodeﬁciency candidate genes

Table 1 | New PID genes predicted from the list of novel PID-candidate genes.

Predicted Known Biological distance Rank of p-Value (percentile) Route between PID gene PID gene between candidate candidate of candidate in candidate and candidate candidate and known in known known known

BCL10 CARD11 1.001 1 0.00007 CARD11 ↔ BCL10 IRF7 MYD88 1.001 1 0.00007 MYD88 ↔ IRF7 IL21 IL21R 1.001 1 0.00007 IL21R ↔ IL21 CTLA4 ICOS 1.616 3 0.00021 ICOS ↔ CTLA4 STING (TMEM173) TBK1 1.001 3 0.00021 TBK1 ↔TMEM173 NFKB1 NFKBIA 1.001 4 0.00028 NFKBIA ↔ NFKB1 NLRC4 NOD2 1.183 5 0.00035 NOD2 ↔ NLRC4 NIK (MAP3K14) CD40 1.064 8 0.00057 CD40 ↔ MAP3K1 TPP1 TINF2 1.191 18 0.00127 TINF2 ↔TPP1 JAGN1 (JAG1) CHD7 2.387 27 0.00191 CHD7 ↔ JAG1 TGFBR2 C8B 6.441 27 0.00191 C8B ↔ CLU ↔TGFBR2 TGFBR1 C8B 6.441 28 0.00198 C8B ↔ CLU ↔TGFBR1 INO80 ACTB 1.582 35 0.00248 ACTB ↔ INO80 DOCK2 RAC2 1.610 35 0.00248 RAC2 ↔ DOCK2 STAT4 STAT3 1.104 35 0.00248 STAT3 ↔ STAT4 ADA2 (TADA2A) TBX1 7.228 83 0.00587 TBX1 ↔ C11orf30 ↔TADA2A IFIH1 FADD 4.141 87 0.00616 FADD ↔ MAVS ↔ IFIH1

Showing the 17 (of 21) PID genes recently reported and identified in the list of PID-candidate genes described in this study. of 21 newly discovered PID genes were present in our proposed an extended list of cancer gene candidates could potentially be list of PID-candidate genes. We plan to use this list, together with identified by HGC-predicted biological distance analysis of known the list of 229 known PID genes, in investigations of the HTS cancer genes, to generate a final list of predicted novel candidates data for PID patients. A hit for one of these genes in this analysis on the basis of biological function relevance. We stress that the would be associated with a higher likelihood of the gene being predicted gene candidates (for PID and other diseases) should not PID-causing. be used for the purpose of excluding irrelevant genes, but rather It is important to note that as in any high-throughput genome- to help investigators to identify novel disease-causing candidate wide analysis, the choice of input data, algorithm, and even small genes. Moreover, as in any other in silico prediction methodology, fine-tuning is likely to strongly affect the outcome. However, we in order for a mutation in a candidate gene to be confirmed as attempted here to provide a reliable prediction where false neg- disease-causing, it must be verified by experimental immunolog- atives (i.e., true PID genes that are not in the final provided list ical and genetic approaches, due to the complex nature of genetic of candidate genes) are minimized – an aim that is expected to disease pathogenesis, which in many cases involves phenotypic be achieved by following a biologically plausible hypothesis of heterogeneity and incomplete penetrance, which current in sil- the tight intra-relatedness of PID genes, together with having a ico methods cannot predict. We believe that rigorous use of the small number of false positives, achieved by estimating the bio- extended in silico-predicted list of candidate PID genes would logical relevance of the gene candidates. The main limitation of increase the rate of novel PID-gene discovery in high throughput the approach described in this study is that it is still expected to sequencing studies (40). contain a large number of false positives: although the final list of 3,110 genes removes about 85% of irrelevant protein-coding AUTHOR CONTRIBUTIONS genes, the function of a large proportion of human genes is poorly YI initiated the study, analyzed the data, and wrote the article. JC understood. Future advances in characterizing the functions of supervised the study, assisted with data resources, and wrote the human genes and updates to genome-wide curated databases of article. gene ontologies and biological/genetic pathways should significantly improve the predictive power of our (and other in silico) ACKNOWLEDGMENTS described methodology. We thank Laurent Abel, Capucine Picard, Bertrand Boisson, Shen- The procedures described here could be used to infer novel Ying Zhang, Stephanie Boisson-Dupuis, Anne Puel, Jacinta Busta- disease gene candidates for other disease groups. For example, as mante, Emmanuelle Jouanguy, Xiao-Fei Kong, Janet Markle, and information about known cancer driver genes is available (39)4, Guillaume Vogt for sharing known and novel PID gene data and advice. We thank Yelena Nemirovskaya and Eric Anderson for administrative support. YI was supported in part by grant # UL1 4http://cancer.sanger.ac.uk/cancergenome/projects/cosmic/ TR000043 from the National Center for Advancing Translational

Frontiers in Immunology | Primary Immunodeﬁciencies April 2015 | Volume 6 | Article 142| 6 Itan and Casanova Novel primary immunodeﬁciency candidate genes

Sciences (NCATS), National Institutes of Health (NIH) Clinical 13. Franceschini A, Szklarczyk D, Frankild S, Kuhn M, Simonovic M, Roth and Translational Science Award (CTSA) program. This study A, et al. STRING v9.1: protein-protein interaction networks, with increased was supported by the Rockefeller University and the St. Giles coverage and integration. Nucleic Acids Res (2013) 41:D808–15. doi:10.1093/ nar/gks1094 Foundation. 14. Hagberg AA, Schult DA, Swart PJ. Exploring Network Structure, Dynamics, and Function Using NetworkX. In: Varoquaux G, Vaught T, Millman J, editors. SUPPLEMENTARY MATERIAL Proceedings of the 7th Python in Science Conference (SciPy2008). Pasadena, CA The Supplementary Material for this article can be found (2008). p. 11–5. online at http://journal.frontiersin.org/article/10.3389/fimmu. 15. Fruchterman TMJ, Reingold EM. Graph drawing by force-directed placement. 2015.00142 Softw Pract Exp (1991) 21:1129–64. 16. Mousallem T, Yang J, Urban TJ, Wang H, Adeli M, Parrott RE, et al. A non- sense mutation in IKBKB causes combined immunodeficiency. Blood (2013) Table S1 | Functional enrichment of known PID genes. Biological GO 124:2046–50. doi:10.1182/blood-2014-04-571265 enrichment categories (p < 0.05) for the 229 known PID genes. Other 17. Willmann KL, Klaver S, Dogu F, Santos-Valente E, Garncarz W, Bilic I, et al. categories include GO accession term and number, fold-enrichment for the PID Biallelic loss-of-function mutation in NIK causes a primary immunodeficiency gene set for the specific GO term, and set of known PID genes for the specific with multifaceted aberrant lymphoid immunity. Nat Commun (2014) 5:5360. GO term. doi:10.1038/ncomms6360 18. Torres JM, Martinez-Barricarte R, García-Gómez S, Mazariegos MS, Itan Table S2 | Proposed novel candidate PID genes. The end result of this study is Y, Boisson B, et al. Inherited BCL10 deficiency impairs hematopoietic and shown in 3,110 novel candidate PID-causing genes predicted in silico (column nonhematopoietic immunity. J Clin Invest (2014) 124:5239–48. doi:10.1172/ A). Column B corresponds to the known PID gene closest to the predicted novel JCI77493 PID gene. Other categories, by column: (C) biological distance between the 19. Martin E, Palmic N, Sanquer S, Lenoir C, Hauck F, Mongellaz C, et al. CTP syn- known and predicted PID genes; (D) the rank of the predicted PID gene in the thase 1 deficiency in humans reveals its central role in lymphocyte proliferation. connectome of the known PID gene; (E) p-value for biological proximity of the Nature (2014) 510:288–92. doi:10.1038/nature13386 predicted PID gene in the connectome of the known PID gene; (F) the predicted 20. Chakraborty PK, Schmitz-Abe K, Kennedy EK, Mamady H, Naas T, Durie D, route (i.e., the genes on the shortest predicted route) between the known and et al. Mutations in TRNT1 cause congenital sideroblastic anemia with immun- predicted PID genes; and (G) the degrees of separation (i.e., the number of odeficiency, fevers, and developmental delay (SIFD). Blood (2014) 124:2867–71. direct connections on the route) between the known and predicted doi:10.1182/blood-2014-08-591370 PID genes. 21. Kocak H, Ballew BJ, Bisht K, Eggebeen R, Hicks BD, Suman S, et al. Hoyeraal- Hreidarsson syndrome caused by a germline mutation in the TEL patch of REFERENCES the telomere protein TPP1. Genes Dev (2014) 28:2090–102. doi:10.1101/gad. 1. Goldstein DB, Allen A, Keebler J, Margulies EH, Petrou S, Petrovski S, et al. 248567.114 Sequencing studies in human genetics: design and interpretation. Nat Rev Genet 22. Kracker S, Di Virgilio M, Schwartzentruber J, Cuenin C, Forveille M, Deau MC, (2013) 14:460–70. doi:10.1038/nrg3455 et al. An inherited immunoglobulin class-switch recombination deficiency asso- 2. Metzker ML. Sequencing technologies – the next generation. Nat Rev Genet ciated with a defect in the INO80 chromatin remodeling complex. J Allergy Clin (2010) 11:31–46. doi:10.1038/nrg2626 Immunol (2014). doi:10.1016/j.jaci.2014.08.030 3. Bamshad MJ, Ng SB, Bigham AW, Tabor HK, Emond MJ, Nickerson DA, et al. 23. Zhang Q, Ji XW,Hou XM, Lu FM, Du Y,Yin JH, et al. Effect of functional nuclear Exome sequencing as a tool for Mendelian disease gene discovery. Nat Rev Genet factor-kappaB genetic polymorphisms on hepatitis B virus persistence and their (2011) 12:745–55. doi:10.1038/nrg3031 interactions with viral mutations on the risk of hepatocellular carcinoma. Ann 4. Casanova JL, Abel L. The genetic theory of infectious diseases: a brief history Oncol (2014) 25:2413–9. doi:10.1093/annonc/mdu451 and selected illustrations. Annu Rev Genomics Hum Genet (2013) 14:215–43. 24. Schubert D, Bode C, Kenefeck R, Hou TZ, Wing JB, Kennedy A, et al. Autosomal doi:10.1146/annurev-genom-091212-153448 dominant immune dysregulation syndrome in humans with CTLA4 mutations. 5. Itan Y,Zhang SY,Vogt G, Abhyankar A, Herman M, Nitschke P,et al. The human Nat Med (2014) 20:1410–6. doi:10.1038/nm.3746 gene connectome as a map of short cuts for morbid allele discovery. Proc Natl 25. Salzer E, Kansu A, Sic H, Májek P, Ikinciogullari A, Dogu FE, et al. Early- Acad Sci U S A (2013) 110:5558–63. doi:10.1073/pnas.1218167110 onset inflammatory bowel disease and common variable immunodeficiency-like 6. Al-Herz W, Bousfiha A, Casanova JL, Chatila T, Conley ME, Cunningham- disease caused by IL-21 deficiency. J Allergy Clin Immunol (2014) 133(1651– Rundles C, et al. Primary immunodeficiency diseases: an update on the classifi- 1659):e1612. doi:10.1016/j.jaci.2014.02.034 cation from the international union of immunological societies expert commit- 26. Kuehn HS, Ouyang W, Lo B, Deenick EK, Niemela JE, Avery DT, et al. Immune tee for primary immunodeficiency. Front Immunol (2014) 5:162. doi:10.3389/ dysregulation in human subjects with heterozygous germline mutations in fimmu.2014.00162 CTLA4. Science (2014) 345:1623–7. doi:10.1126/science.1255904 7. Casanova JL, Conley ME, Seligman SJ, Abel L, Notarangelo LD. Guidelines for 27. Jeremiah N, Neven B, Gentili M, Callebaut I, Maschalidi S, Stolzenberg MC, genetic studies in single patients: lessons from primary immunodeficiencies. J et al. Inherited STING-activating mutation underlies a familial inflammatory Exp Med (2014) 211:2137–49. doi:10.1084/jem.20140520 syndrome with lupus-like manifestations. J Clin Invest (2014) 124:5516–20. 8. Itan Y, Mazel M, Mazel B, Abhyankar A, Nitschke P, Quintana-Murci L, et al. doi:10.1172/JCI79100 HGCS: an online tool for prioritizing disease-causing gene variants by biological 28. Rice GI, del Toro Duany Y, Jenkinson EM, Forte GM, Anderson BH, Ariaudo G, distance. BMC Genomics (2014) 15:256. doi:10.1186/1471-2164-15-256 et al. Gain-of-function mutations in IFIH1 cause a spectrum of human disease 9. Flicek P, Ahmed I, Amode MR, Barrell D, Beal K, Brent S, et al. Ensembl 2013. phenotypes associated with upregulated type I interferon signaling. Nat Genet Nucleic Acids Res (2013) 41:D48–55. doi:10.1093/nar/gks1236 (2014) 46:503–9. doi:10.1038/ng.2933 10. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis 29. Zhou Q, Yang D, Ombrello AK, Zavialov AV, Toro C, Zavialov AV, et al. Early- of large gene lists using DAVID bioinformatics resources. Nat Protoc (2009) onset stroke and vasculopathy associated with mutations in ADA2. N Engl J Med 4:44–57. doi:10.1038/nprot.2008.211 (2014) 370:911–20. doi:10.1056/NEJMoa1307361 11. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene 30. Navon Elkan P, Pierce SB, Segel R, Walsh T, Barash J, Padeh S, et al. Mutant ontology: tool for the unification of biology. The gene ontology consortium. adenosine deaminase 2 in a polyarteritis nodosa vasculopathy. N Engl J Med Nat Genet (2000) 25:25–9. doi:10.1038/75556 (2014) 370:921–31. doi:10.1056/NEJMoa1307362 12. Paradis E, Claude J, Strimmer K. APE: analyses of phylogenetics and evolution 31. Ciancanelli MJ, Huang S, Luthra P, Garner H, Itan Y, Volpi S, et al. Life- in R language. Bioinformatics (2004) 20:289–90. doi:10.1093/bioinformatics/ threatening influenza and impaired interferon amplification in human IRF7 btg412 deficiency. Science (2015). (in press).

www.frontiersin.org April 2015 | Volume 6 | Article 142| 7 Itan and Casanova Novel primary immunodeﬁciency candidate genes

32. Dobbs K, Domínguez CC, Zhang SY, Parolini S, Audry M, Chou J, et al. Human 39. Forbes SA, Bindal N, Bamford S, Cole C, Kok CY, Beare D, et al. COSMIC: min- DOCK2 mutations underlie an autosomal recessive pleiotropic immunodefi- ing complete cancer genomes in the catalogue of somatic mutations in cancer. ciency with early-onset invasive infections. N Engl J Med (in press). Nucleic Acids Res (2011) 39:D945–50. doi:10.1093/nar/gkq929 33. Boztug K, Järvinen PM, Salzer E, Racek T, Mönch S, Garncarz W, et al. JAGN1 40. Belkadi A, Bolze A, Itan Y, Cobat A, Vincent QB, Antipenko A, et al. Whole- deficiency causes aberrant myeloid cell homeostasis and congenital neutropenia. genome sequencing is more powerful than whole-exome sequencing for detect- Nat Genet (2014) 46:1021–7. doi:10.1038/ng.3069 ing exome variants. Proc Natl Acad Sci (2015). (in press). 34. Naviglio S, Arrigo S, Martelossi S, Villanacci V, Tommasini A, Loganes C, et al. Severe inflammatory bowel disease associated with congenital alteration of transforming growth factor beta signaling. J Crohns Colitis (2014) 8:770–4. Conflict of Interest Statement: The authors declare that the research was conducted doi:10.1016/j.crohns.2014.01.013 in the absence of any commercial or financial relationships that could be construed 35. Sadat MA, Moir S, Chun TW, Lusso P, Kaplan G, Wolfe L, et al. Glycosylation, as a potential conflict of interest. hypogammaglobulinemia,and resistance to viral infections. N Engl J Med (2014) 370:1615–25. doi:10.1056/NEJMoa1302846 Received: 16 December 2014; accepted: 15 March 2015; published online: 01 April 2015. 36. Sassi A, Lazaroski S, Wu G, Haslam SM, Fliegauf M, Mellouli F, et al. Hypomor- Citation: Itan Y and Casanova J-L (2015) Novel primary immunodeficiency can- phic homozygous mutations in phosphoglucomutase 3 (PGM3) impair immu- didate genes predicted by the human gene connectome. Front. Immunol. 6:142. doi: nity and increase serum IgE levels. J Allergy Clin Immunol (2014) 133:e1–3. 10.3389/fimmu.2015.00142 doi:10.1016/j.jaci.2014.02.025 This article was submitted to Primary Immunodeficiencies, a section of the journal 37. Zhang Y, Yu X, Ichikawa M, Lyons JJ, Datta S, Lamborn IT, et al. Autosomal Frontiers in Immunology. recessive phosphoglucomutase 3 (PGM3) mutations link glycosylation defects Copyright © 2015 Itan and Casanova. This is an open-access article distributed under to atopy, immune deficiency, autoimmunity, and neurocognitive impairment. the terms of the Creative Commons Attribution License (CC BY). The use, distribution J Allergy Clin Immunol (2014) 133:e1–5. doi:10.1016/j.jaci.2014.02.013 or reproduction in other forums is permitted, provided the original author(s) or licensor 38. Aavikko M, Kaasinen E, Nieminen JK, Byun M, Donner I, Mancuso R, et al. are credited and that the original publication in this journal is cited, in accordance with Whole genome sequencing identifies STAT4 as a putative susceptibility gene in accepted academic practice. No use, distribution or reproduction is permitted which classic Kaposi sarcoma. J Infect Dis (2014) pii:jiu667. doi:10.1093/infdis/jiu667 does not comply with these terms.

Frontiers in Immunology | Primary Immunodeﬁciencies April 2015 | Volume 6 | Article 142| 8