bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Single-cell dissection of a rare human prostate basal cell carcinoma

Summary heading：Single-cell analysis of prostate basal cell carcinoma

Abstract

As a rare subtype of prostate carcinoma, basal cell carcinoma (BCC) has not been

studied extensively and thus lacks systematic molecular characterization. Here we

applied single-cell genomic amplification and RNA-Seq to a specimen of human

prostate BCC (CK34βE12+/P63+/PAP-/PSA-). The mutational landscape was obtained

via whole exome sequencing of the amplification mixture of 49 single cells, and the 5

putative driver genes mutated are CASC5, NUTM1, PTPRC, KMT2C and TBX3. The

top 3 nucleotide substitutions are C>T, T>C and C>A, similar to common prostate

cancer. The distribution of the variant allele frequency values indicated these single

cells are from the same tumor clone. The transcriptomes of 69 single cells were

obtained, and they were clustered into tumor, stromal and immune cells based on their

global transcriptomic profiles. The tumor cells specifically express basal cell markers

like KRT5, KRT14 and KRT23, and epithelial markers EPCAM, CDH1 and CD24.

The transcription factor (TF) co-variance network analysis showed that the BCC

tumor cells have distinct regulatory networks. By comparison with current prostate

cancer datasets, we found that some of the bulk samples exhibit basal-cell signatures.

Interestingly, at single-cell resolution the gene expression patterns of prostate BCC

tumor cells show uniqueness compared with that of common prostate cancer-derived

circulating tumor cells. This study, for the first time, discloses the comprehensive

mutational and transcriptomic landscapes of prostate BCC, which lays a foundation

1 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

for the understanding of its tumorigenesis mechanism and provides new insights into

prostate cancers in general.

Xianbin Su1,2, #, Qi Long3, #, Juanjie Bo1, #, Yi Shi2, #, Li-Nan Zhao2, Yingxin Lin4,

Qing Luo2, Shila Ghazanfar4,5, Chao Zhang6, Qiang Liu7, Lan Wang2, Kun-Yan He2,

Jian He2, Xiao-Fang Cui2, Jean Y. H. Yang4, Ze-Guang Han2,*, Jian-Jun Sha1,*,

Guoliang Yang1, *

1Department of Urology and Key Laboratory of Systems Biomedicine (Ministry of

Education), Renji Hospital, School of Medicine, Shanghai Jiao Tong University,

Shanghai, China

2Key Laboratory of Systems Biomedicine (Ministry of Education) and Collaborative

Innovation Center of Systems Biomedicine, Shanghai Center for Systems

Biomedicine, Shanghai Jiao Tong University, Shanghai, China

3Key Laboratory for Regenerative Medicine (Ministry of Education), School of

Biomedical Sciences, Faculty of Medicine, The Chinese University of Hong Kong,

Hong Kong SAR, China

4Charles Perkins Center and School of Mathematics and Statistics, The University of

Sydney, Sydney, Australia

5Cancer Research UK Cambridge Institute, University of Cambridge, Li Ka Shing

Centre, Robinson Way, Cambridge CB2 0RE, United Kingdom

6Department of Urology, Shanghai Changhai Hospital, Second Military Medical

2 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

University, Shanghai, China

7Department of Pathology, Renji Hospital, School of Medicine, Shanghai Jiao Tong

University, Shanghai, China

#These authors contributed equally.

*To whom correspondence should be addressed:

Guoliang Yang, Department of Urology, Renji Hospital, School of Medicine,

Shanghai Jiao Tong University, Shanghai 200127, China. Tel: +86-21-6373 0455; Fax:

+86-21-6373 0455; Email: [email protected]

Jian-Jun Sha, Department of Urology, Renji Hospital, School of Medicine, Shanghai

Jiao Tong University, Shanghai 200127, China. Tel: +86-21-6373 0455; Fax:

+86-21-6373 0455; Email: [email protected]

Ze-Guang Han, Key Laboratory of Systems Biomedicine (Ministry of Education),

Shanghai Center of Systems Biomedicine, Shanghai Jiao Tong University, Shanghai

200240, China. Tel: +86-21-3420 7304; Fax: +86-21-3420 6059; Email:

[email protected]

3 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Running title

Single-cell analysis of prostate basal cell carcinoma

Abbreviations BCC, basal cell carcinoma; CTCs, circulating tumor cells; MDA, multiple

displacement amplification; PAP, prostatic acid phosphatase; PSA, prostate-specific

antigen; TF, transcription factor; VAF, variant allele frequency.

Keywords

prostate, basal cell carcinoma, single-cell sequencing, gene expression, transcription

factor, mutation

Conflict of Interest

The authors declare that they have no conflicts of interest with the contents of this

article.

4 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Introduction

Prostate cancer is a heterogeneous disease with complex subtypes and different cell

origins [1-3], and basal cell carcinoma (BCC) is an extremely rare histological

subtype comprising < 0.01% of prostate cancer [4]. Despite being largely regarded as

pursuing an indolent clinical course, aggressive behaviors like recurrence or

metastasis have been observed and deaths also reported, highlighting the complex and

yet poorly understood mechanism [5-9]. There have been extensive reports on the

genomic features of common prostate cancer [10-17], and the molecular features of

normal human prostate basal cells, prostate basal cell hyperplasia, and basal

populations from human prostate cancer have also been reported [18-20], but the

molecular profile of prostate BCC is still lacking. The mutational and transcriptomic

features of this tumor sub-type will thus be valuable for understanding of its

tumorigenesis mechanism and development of future clinical treatments.

Single-cell sequencing has evolved to be a powerful tool that provides

unprecedented resolution of clinical specimens especially with limited amounts, and

marker-free decomposition of the constituent cell types based on single-cell

transcriptional profiles allows precise dissection of complex systems such as various

tissues or tumors [21-23]. Here we applied single-cell genomic amplification to a

human specimen of prostate BCC and used the mixture of 49 single cells to provide

mutational landscape of prostate BCC via whole exome sequencing. We then used

single-cell RNA-Seq to obtain the transcriptomes of 69 cells from the same specimen.

Three types of cells, namely tumor, stromal and immune cells, were identified with

5 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

their molecular features revealed at single-cell resolution, providing a useful resource

for not only this prostate cancer subtype but prostate cancer in general.

Results Histological and immunohistochemical features of human prostate BCC

The tumor exhibited a widespread infiltrative growth pattern under microscopic

examination, and individual cells had large pleomorphic nuclei and scant cytoplasm

(Fig. 1). The tumor was negative for two known prostatic markers, prostatic acid

phosphatase (PAP) and prostate-specific antigen (PSA), but strongly expressed

CK34βE12 and P63, suggestive of basal cell origin (Fig. 1). Based on the cellular

morphology and immunohistochemical staining, we diagnosed it as prostate BCC.

The patient with BCC was resistant to chemotherapy but sensitive to radiotherapy,

and followed-up for 32 months without recurrence or metastasis (Supporting

Information, Fig. S1).

Whole exome sequencing reveals the mutational landscape of prostate BCC

As genetic variations are generally believed to be the causes of tumors, here we tried

to reveal the mutational features of prostate BCC. Due to the limited amount of tumor

specimens collected, after dissociation of the specimens into single-cell suspension,

we first conducted single-cell whole genome multiple displacement amplification

(MDA) on a microfluidic-chip. We then used the mixture of amplified products from

49 single cells for whole exome sequencing (Fig. 2A). A total of 91 functional exonic

mutations were obtained, and 5 of the mutated genes, CASC5, NUTM1, PTPRC,

6 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

KMT2C and TBX3 are putative driver genes catalogued in the COSMIC Cancer Gene

Census (Supporting Information, Fig. S2) [24]. Mutations in DNA repair genes are

related to poor outcomes of prostate cancers [12, 25], but few are mutated in this

prostate BCC sample such as BRCA2, consistent with good survival of this patient.

The top 3 nucleotide substitutions are C>T, T>C and C>A, which are exactly the

same top 3 substitutions in recently reported localized, non-indolent prostate tumors

(Fig. 2B) [13]. This suggests the etiology of prostate BCC is likely the same as

common prostate tumors. The major composite COSMIC signatures are Signature 4,

5, 6 and 12 (Fig. 2C). While the etiologies of Signature 5 and 12 are still unknown,

Signature 4 and 6 are associated with smoking and defective DNA mismatch repair,

respectively [26].

By checking the distribution of the variant allele frequency (VAF) of the exonic

mutations of prostate BCC, it is clear that most of the variations have VAF values of

around 0.5, thus supporting that most of the single cells share a similar mutational

background (Fig. 2D). The VAF values of the 5 driver mutations are also

approximately 0.5 (Fig. 2E), indicating these single cells are probably originated from

the same ancestor and form the major tumor clone. However, there are still some

variations with lower VAF values which suggest sub-clonal mutations may have

existed in some cells that are acquired during later diversified evolution of the tumor.

To understand whether the prostate BCC is showing similarities to other reported

prostate cancer cases, we checked the mutational frequency of the prostate BCC

mutated genes in prostate cancer samples collected in cBioPortal [27]. For the 88

7 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

genes enquired, only 10 genes are not mutated, and 27 genes are mutated in more than

0.5% of the ~4400 prostate cancer samples (Fig. 2F). Interestingly, 3 of the driver

genes from this study, KMT2C, PTPRC, and TBX3 are also among the top mutated

genes, especially KMT2C with mutation rate of ~6%. GSEA/MSigDB enrichment

analysis [28] of the mutated genes shows that the top enriched terms are related to cell

cycle, cytoskeleton and others (Fig. 2G), consistent with its basal cell origin.

Single-cell RNA-Seq reveals the transcriptomic features of prostate BCC

To understand the transcriptional profiles of prostate BCC, we then randomly selected

69 single cells for single-cell RNA-Seq (Fig. 3A). With a median unique mapping rate

of 58.4% and 1.08 million mapped reads, we obtained a median of ~2800 genes per

cell (Supporting Information, Fig. S3A). We used ERCC spike-ins as positive controls,

and the high correlation efficiency between single-cells based on the 92 spike-ins

confirmed the high quality of the data (Supporting Information, Fig. S3B-D). After

filtering 5 outliers, the remaining 64 cells were classified into three clusters based on

their global transcriptomic profiles (Fig. 3B-C). GSEA/MSigDB enrichment analysis

identified them as tumor cells (51 cells), stromal cells (7 cells) and immune cells (6

cells) (Fig. 3D).

The three clusters of cells can also be clearly separated in tSNE plot based on the

top genes expressed in each cluster (Fig. 4A). The tumor cells specifically expressed

genes encoding cytokeratins including KRT5, KRT14 and KRT23, which are reliable

basal cell markers (Fig. 4B). Interestingly, epithelial markers EPCAM, CDH1 and

8 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

CD24 are also highly expressed in tumor cells. Stromal cells specifically express

COL1A2, THY1 and BGN, and immune cells specifically express monocytic marker

CD14, CXCL8 and myeloid cell nuclear differentiation antigen MNDA (Fig. 4B),

suggestive of possible innate immune cell infiltration. Immunohistochemical staining

confirmed the expression of CK14 (KRT14) at protein level, further proving the

reliability of the sequencing data (Fig. 4C).

Transcription factors (TFs) covariance network analysis showed that the three

clusters demonstrate different transcriptional regulatory status (Fig. 4D). Significantly,

MYC, ARID1A, TBX3 and SOX11 are highly expressed in tumor cells, where, besides

the well-known MYC that is recently reported to be over-expressed in human prostatic

basal cells [19], the oncogenic role of ARID1A, a key component of chromatin

remodeling complex SWI/SNF, has been recently demonstrated in liver cancer [29],

the development-related gene SOX11 was reported to be elevated in basal-like breast

cancer [30], and TBX3 plays important roles in development, differentiation and

tumorigenesis [31]. The TFs co-variance network provides clues to future dissection

of the regulatory mechanism of BCC.

Single-cell analysis of prostate BCC provides new insights to prostate cancer

We then checked the expression of prostate BCC tumor cell specific genes in

previously reported luminal and basal cells sorted from normal and cancerous

prostates by flow cytometry [18]. The prostate basal cells exhibit higher expression

levels of KRT5, KRT14 and KRT23 compared with luminal cells, similarly as our

9 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

prostate BCC tumor cells, further confirming their basal cell origin (Fig. 5A).

Interestingly, the genes highly expressed in prostate BCC tumor cells fall into two

groups based on their expression correlation patterns, with one group including KRT5,

KRT14, KRT23, DSC3, FGFR2, etc., while the other group including EPCAM, CDH1,

CD24, etc. (Fig. 5A). It seems the former group includes the basal cell markers while

the latter group includes genes expressed in common prostate epithelial cells.

We further checked the expression patterns of BCC tumor cell specific genes in

other bulk samples of human prostate cancers [10, 11, 32]. It is interesting to identify

a small group of samples exhibiting basal cell gene expression features in two

separate prostate cancer datasets (shaded branches in the dendrograms of Fig. 5B and

Supporting Information, Fig. S4A). Likely explanations are that these tumor samples

were derived from BCC patients or the samples contained basal cells. For TCGA

PRAD dataset, the expressions of basal markers are detected in most samples,

possibility due to the high number of genes detected by deep sequencing (Supporting

Information, Fig. S4B). The split of the genes highly expressed in BCC tumor cells

into 2 groups is further confirmed by these three datasets, with KRT5, KRT14, KRT23

clustering together with DSC3, FGFR2 and also possibly ACTG2. Our single-cell

analysis of prostate BCC reveals the existence of basal cell features in current bulk

samples of prostate cancer.

We then compared our results with one single-cell resolution dataset of prostate

cancer [33]. Compared with circulating tumor cells (CTCs) from prostate cancer

patients, the prostate BCC tumor cells exhibit distinct gene expression features with

10 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

high levels of IGF2, KRT5, KRT14, KRT23, AQP1, MIA, FGFR2 and ACTG2,

although they all expressed EPCAM, CDH1 and CD24 (Fig. 5C). The results

demonstrated the unique gene expression patterns of the prostate BCC tumor cells,

which exhibit the molecular features of basal cells with activated genes like IGF2 and

FGFR2, consistent with report on the regulation of pluripotency by IGF and FGF

pathways [34]. Another interesting finding is that the BCC immune cells clustered

together with some single cells annotated as CTC-candidate, which were likely to be

contaminating leukocytes to the CTCs (Fig. 5C). The single-cell level data also

support the above mentioned split of the genes highly expressed in prostate BCC

tumor cells, with one group representing genuine BCC markers while the other genes

commonly expressed in prostate cancer.

Detection of exonic mutations in single-cell RNA-Seq reads

As our mutational data and single-cell RNA-Seq data were generated from the same

prostate BCC specimen, we checked whether the exonic mutations could be detected

in RNA-Seq reads. Interestingly, we indeed found mutations in RNA-Seq reads from

some single cells for GABPB2, ASTE1, COPS3 and BEX2 (Supporting Information,

Fig. S5). The variations are only detected in some of the single cells, which are likely

because the mutated sites are not always covered by the relatively shallow single-cell

RNA-Seq. Another phenomenon is that single cells in RNA-Seq analysis only

contained either reference or altered variant, which is probably caused by allelic

11 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

specific expression or allele dropout in RNA-Seq. The consistency of scRNA-Seq

data and exome-seq data further suggested the reliability of our prostate BCC data.

Discussion As a rare malignancy, there has been little research into the molecular characteristics

of prostate BCC and no consensus on its treatment. Mostly, such data exist as

sporadic reports on immunohistochemical expression in a small number of cases [8,

35-37]. This study represents the first genomic and transcriptomic profiling of this

prostate cancer subtype, and the single-cell resolution transcriptomic data is especially

valuable for understanding of not only this subtype but prostate cancer in general.

The molecular features of normal human prostate basal cells [19], prostate basal

cell hyperplasia [20], and human prostate cancer derived basal populations [18] have

been reported recently, and the single-cell transcriptomic profiles of prostate BCC is a

good complement to these data for comparison studies and better understanding of

prostate cancer cell origin. Most of the current genomics data for prostate cancer are

generated using bulk samples [10, 11, 13], and the single-cell transcriptional profiles

of prostate BCC provide an opportunity for deeper utilization of these data. For

example, analysis using genes highly expressed in prostate BCC reveals the existence

of basal cell features in some bulk samples of prostate cancer, which is consistent

with recent classification of prostate cancer samples into basal-like and luminal-like

subtypes using the PAM50 classifier [1, 38]. The single-cell resolution profiles of

constitutional cell types from prostate BCC provide further advantage for future

12 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

de-convolution of bulk samples [39], which will maximize the values of current

genomic data of prostate cancer.

The prostate BCC patient is still surviving now, indicating this case is an indolent

malignancy. Our genomics and transcriptomic analysis of this chemotherapy-resistant

but radiotherapy-sensitive prostate BCC provides a useful resource for future

dissection of the molecular mechanisms related to the radiotherapy responsiveness

and treatment guidance. Mutations that disrupt the function of DNA damage repair

genes have been shown to be associated with the aggressive clinical behavior of

localized prostate cancer and with cancer-specific mortality [12, 40-42], and it has

also been reported that men with metastatic prostate cancer and DNA-repair gene

mutations have sustained responses to poly-ADP ribose polymerase (PARP)

inhibitors and platinum-based chemotherapy [43, 44]. This study showed that the

prostate BCC patient with few DNA-repair gene mutations was resistant to

chemotherapy, suggesting that we may directly adopt radiotherapy rather than

chemotherapy for such conditions. The results indicated that precision treatment of

BCC could be more than just a histopathological triviality but be based on association

of tumor phenotypes with molecular features.

In summary, single-cell analysis revealed the genomic and transcriptomic

landscapes of the rare prostate BCC for the first time. This provides clues for

elucidation of the tumorigenesis mechanism, further discovery of biomarkers and

therapeutic targets, and better understanding of not only prostate BCC but prostate

cancer in general.

13 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Methods

Clinical specimens

This study was approved by the Ethnical Review Board of Renji Hospital, School of

Medicine, Shanghai Jiao Tong University. Specimens were obtained by prostatic

needle biopsy from a 55-year-old man with pelvic pain and irritative urinary

symptoms. The specimens underwent pathological evaluation. After diagnosis as

prostate BCC, the patient was treated with chemotherapy and radiotherapy and then

followed-up for 32 months without evidence of local recurrence or distant metastases

(Supporting Information, Fig. S1). The specimens were mechanically dissociated and

digested with collagenase IV and DNase I, and the single-cell suspensions were used

for both single-cell genome amplification and single-cell RNA-Seq.

Immunohistochemical staining

Immunohistochemical staining was done as previously described [45], and we used

the following primary antibodies: P63 (abcam, ab124762), CK34βE12 (Dako,

M0630), CK14 (abcam, ab7800), PAP (Dako, M0792), PSA (Dako, A056201). We

used biotinylated universal link antibody as the secondary antibody.

Single-cell genome amplification and whole exome sequencing

Single-cell capture, lysis, reverse-transcription, and whole genome MDA were done

in a microfluidic-based C1 DNA-Seq IFC (10~17 μm, Fluidigm) according to its

14 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

protocol using illustra GenomiPhi V2 DNA Amplification Kit (GE Healthcare,

25660031). The amplified products from 49 single cells were mixed for exonic region

capture and library preparation using Agilent SureSelect Human All Exon v7 Kit

(Agilent, 5191-4005). The library was sequenced using illumina HiSeq platform with

2 × 150 bp sequencing mode. We used GATK [46] for variant calling and ANNOVAR

[47] for functional annotation of the mutations. As there is no para-tumor tissue for

this patient, we filtered SNPs using dbSNP141 [48] and 1,000 Genomes Project (v3)

database [49]. We used MutationalPatterns [50] to decipher the mutational signature

composition. For genes mutated in this prostate BCC sample, we checked their

mutational frequencies in prostate cancer samples collected in cBioPortal [27].

GSEA/MSigDB enrichment analysis was also conducted for the genes mutated in this

prostate BCC.

Single-cell RNA-Seq and data analysis

Single-cell RNA-Seq experiment and analysis were conducted as previously

described [45]. Single-cell capture, lysis, reverse-transcription, and cDNA

amplification were done in a microfluidic-based C1 RNA-Seq IFC (10~17 μm,

Fluidigm). After library preparation with Nextera XT Kit and Index Kit (illumina), the

single-cell libraries were pooled and sequenced using NextSeq 500 (illumina) with 2

× 76 bp sequencing mode. After generation of FPKM data, SC3 [51] was used for

identification of cell outliers and single-cell clustering. The genes differently

expressed between clusters were then identified via ANOVA analysis (p < 0.05) using

15 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

SINGuLAR™ (Fluidigm). GSEA/MSigDB enrichment analysis was conducted for

the genes specifically expressed in each cluster to facilitate cell type identification.

Transcription factors (TFs) covariance network analysis was conducted to reveal the

relationship of the TFs specifically expressed in each cell cluster as previously

described [45, 52]. Gene expression files were retrieved from five other prostate

cancer studies [10, 11, 18, 32, 33] for comparison studies.

Acknowledgement This work is supported in part by the National Natural Science Foundation of China

(81802806, 81472621, 81402329 and 81902561), National Program on Key

Research Project of China (2016YFC0902701, Precision Medicine), Shanghai Health

Bureau (20164Y0124), Medical and Engineering Crossover Fund of SJTU

(YG2016QN71, YG2017MS67), University of Sydney – Shanghai Jiao Tong

University Joint Research Alliance (USYD-SJTU JRA) grant, funding from Key

Laboratory of Systems Biomedicine (Ministry of Education) (KLSB2017QN-03), and

Incubating Program for Clinical Research and Innovation of Renji Hospital SJTU

School of Medicine (PYII-17-010).

Authors' contributions

XS and GY designed the study. XS, JB, LZ, CZ, QL3, LW, KH, JH, XC, WX and GY

carried out experiments. XS, QL1, YS, YL, QL2, SG and JY analyzed data. XS, GY

and ZH drafted the manuscript. GY, JS and ZH supervised the study. All authors read

and approved final version of the manuscript. 16 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

References

1. Zhao SG, Chang SL, Erho N, Yu M, Lehrer J, Alshalalfa M, Speers C, Cooperberg MR, Kim W, Ryan CJ, et al. (2017) Associations of Luminal and Basal Subtyping of Prostate Cancer With Prognosis and Response to Androgen Deprivation Therapy. JAMA Oncol 3, 1663-1672.

2. Goldstein AS, Huang J, Guo C, Garraway IP & Witte ON (2010) Identification of a cell of origin for human prostate cancer. Science 329, 568-571.

3. Choi N, Zhang B, Zhang L, Ittmann M & Xin L (2012) Adult murine prostate basal and luminal cells are self-sustained lineages that can both serve as targets for prostate cancer initiation. Cancer Cell 21, 253-265.

4. Tannenbaum M (1975) Adenoid cystic or "salivary gland" carcinomas of prostate. Urology 6, 238-239.

5. Halat SK & MacLennan GT (2008) Adenoid cystic/basal cell carcinoma of the prostate. J Urol 179, 1576.

6. Komura K, Inamoto T, Tsuji M, Ibuki N, Koyama K, Ubai T, Azuma H & Katsuoka Y (2010) Basal cell carcinoma of the prostate: unusual subtype of prostatic carcinoma. Int J Clin Oncol 15, 594-600.

7. Chang K, Dai B, Kong Y, Qu Y, Wu J, Ye D, Yao X, Zhang S, Zhang H, Zhu Y, et al. (2013) Basal cell carcinoma of the prostate: clinicopathologic analysis of three cases and a review of the literature. World J Surg Oncol 11, 193.

8. Ali TZ & Epstein JI (2007) Basal cell carcinoma of the prostate: a clinicopathologic study of 29 cases. Am J Surg Pathol 31, 697-705.

9. Ayyathurai R, Civantos F, Soloway MS & Manoharan M (2007) Basal cell carcinoma of the prostate: current concepts. BJU Int 99, 1345-1349.

10. The Cancer Genome Atlas Research Network (2015) The Molecular Taxonomy 17 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

of Primary Prostate Cancer. Cell 163, 1011-1025.

11. Robinson D, Van Allen EM, Wu YM, Schultz N, Lonigro RJ, Mosquera JM, Montgomery B, Taplin ME, Pritchard CC, Attard G, et al. (2015) Integrative clinical genomics of advanced prostate cancer. Cell 161, 1215-1228.

12. Pritchard CC, Mateo J, Walsh MF, De Sarkar N, Abida W, Beltran H, Garofalo A, Gulati R, Carreira S, Eeles R, et al. (2016) Inherited DNA-Repair Gene Mutations in Men with Metastatic Prostate Cancer. N Engl J Med 375, 443-453.

13. Fraser M, Sabelnykova VY, Yamaguchi TN, Heisler LE, Livingstone J, Huang V, Shiah YJ, Yousif F, Lin X, Masella AP, et al. (2017) Genomic hallmarks of localized, non-indolent prostate cancer. Nature 541, 359-364.

14. Grasso CS, Wu YM, Robinson DR, Cao X, Dhanasekaran SM, Khan AP, Quist MJ, Jing X, Lonigro RJ, Brenner JC, et al. (2012) The mutational landscape of lethal castration-resistant prostate cancer. Nature 487, 239-243.

15. Quigley DA, Dang HX, Zhao SG, Lloyd P, Aggarwal R, Alumkal JJ, Foye A, Kothari V, Perry MD, Bailey AM, et al. (2018) Genomic Hallmarks and Structural Variation in Metastatic Prostate Cancer. Cell 174, 758-769.

16. Espiritu SMG, Liu LY, Rubanova Y, Bhandari V, Holgersen EM, Szyca LM, Fox NS, Chua MLK, Yamaguchi TN, Heisler LE, et al. (2018) The Evolutionary Landscape of Localized Prostate Cancers Drives Clinical Aggression. Cell 173, 1003-1013 e1015.

17. Baca SC, Prandi D, Lawrence MS, Mosquera JM, Romanel A, Drier Y, Park K, Kitabayashi N, MacDonald TY, Ghandi M, et al. (2013) Punctuated evolution of prostate cancer genomes. Cell 153, 666-677.

18. Smith BA, Sokolov A, Uzunangelov V, Baertsch R, Newton Y, Graim K, Mathis C, Cheng D, Stuart JM & Witte ON (2015) A basal stem cell signature identifies

18 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

aggressive prostate cancer phenotypes. Proc Natl Acad Sci U S A 112, E6544-6552.

19. Zhang D, Park D, Zhong Y, Lu Y, Rycaj K, Gong S, Chen X, Liu X, Chao HP, Whitney P, et al. (2016) Stem cell and neurogenic gene-expression profiles link prostate basal cells to aggressive prostate cancer. Nat Commun 7, 10798.

20. Henry G, Malewska A, Mauck R, Gahan J, Hutchinson R, Torrealba J, Francis F, Roehrborn C & Strand D (2017) Molecular pathogenesis of human prostate basal cell hyperplasia. Prostate 77, 1344-1355.

21. Gao S, Yan L, Wang R, Li J, Yong J, Zhou X, Wei Y, Wu X, Wang X, Fan X, et al. (2018) Tracing the temporal-spatial transcriptome landscapes of the human fetal digestive tract using single-cell RNA-sequencing. Nat Cell Biol 20, 721-734.

22. Han X, Wang R, Zhou Y, Fei L, Sun H, Lai S, Saadatpour A, Zhou Z, Chen H, Ye F, et al. (2018) Mapping the Mouse Cell Atlas by Microwell-Seq. Cell 172, 1091-1107 e1017.

23. Tanay A & Regev A (2017) Scaling single-cell genomics from phenomenology to mechanism. Nature 541, 331-338.

24. Sondka Z, Bamford S, Cole CG, Ward SA, Dunham I & Forbes SA (2018) The COSMIC Cancer Gene Census: describing genetic dysfunction across all human cancers. Nat Rev Cancer 18, 696-705.

25. Taylor RA, Fraser M, Rebello RJ, Boutros PC, Murphy DG, Bristow RG & Risbridger GP (2019) The influence of BRCA2 mutation on localized prostate cancer. Nat Rev Urol 16, 281-290.

26. Alexandrov LB, Nik-Zainal S, Wedge DC, Aparicio SA, Behjati S, Biankin AV, Bignell GR, Bolli N, Borg A, Borresen-Dale AL, et al. (2013) Signatures of mutational processes in human cancer. Nature 500, 415-421.

27. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, Sun Y,

19 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Jacobsen A, Sinha R, Larsson E, et al. (2013) Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 6, pl1.

28. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, et al. (2005) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102, 15545-15550.

29. Sun X, Wang SC, Wei Y, Luo X, Jia Y, Li L, Gopal P, Zhu M, Nassour I, Chuang JC, et al. (2017) Arid1a has context-dependent oncogenic and tumor suppressor functions in liver cancer. Cancer Cell 32, 574-589 e576.

30. Shepherd JH, Uray IP, Mazumdar A, Tsimelzon A, Savage M, Hilsenbeck SG & Brown PH (2016) The SOX11 transcription factor is a critical regulator of basal-like breast cancer growth, invasion, and basal-like gene expression. Oncotarget 7, 13106-13121.

31. Wansleben S, Peres J, Hare S, Goding CR & Prince S (2014) T-box transcription factors in cancer biology. Biochim Biophys Acta 1846, 380-391.

32. Beltran H, Prandi D, Mosquera JM, Benelli M, Puca L, Cyrta J, Marotz C, Giannopoulou E, Chakravarthi BV, Varambally S, et al. (2016) Divergent clonal evolution of castration-resistant neuroendocrine prostate cancer. Nat Med 22, 298-305.

33. Miyamoto DT, Zheng Y, Wittner BS, Lee RJ, Zhu H, Broderick KT, Desai R, Fox DB, Brannigan BW, Trautwein J, et al. (2015) RNA-Seq of single prostate CTCs implicates noncanonical Wnt signaling in antiandrogen resistance. Science 349, 1351-1356.

34. Bendall SC, Stewart MH, Menendez P, George D, Vijayaragavan K, Werbowetski-Ogilvie T, Ramos-Mejia V, Rouleau A, Yang J, Bosse M, et al. (2007) IGF and FGF cooperatively establish the regulatory stem cell niche of pluripotent

20 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

human cells in vitro. Nature 448, 1015-1021.

35. Yang XJ, McEntee M & Epstein JI (1998) Distinction of basaloid carcinoma of the prostate from benign basal cell lesions by using immunohistochemistry for bcl-2 and Ki-67. Hum Pathol 29, 1447-1450.

36. Iczkowski KA, Ferguson KL, Grier DD, Hossain D, Banerjee SS, McNeal JE & Bostwick DG (2003) Adenoid cystic/basal cell carcinoma of the prostate: clinicopathologic findings in 19 cases. Am J Surg Pathol 27, 1523-1529.

37. Montironi R, Mazzucchelli R, Stramazzotti D, Scarpelli M, Lopez Beltran A & Bostwick DG (2005) Basal cell hyperplasia and basal cell carcinoma of the prostate: a comprehensive review and discussion of a case with c-erbB-2 expression. J Clin Pathol 58, 290-296.

38. Zhao SG, Chen WS, Das R, Chang SL, Tomlins SA, Chou J, Quigley DA, Dang HX, Barnard TJ, Mahal BA, et al. (2019) Clinical and Genomic Implications of Luminal and Basal Subtypes Across Carcinomas. Clin Cancer Res 25, 2450-2457.

39. Wang X, Park J, Susztak K, Zhang NR & Li M (2019) Bulk tissue cell type deconvolution with multi-subject single-cell expression reference. Nat Commun 10, 380.

40. Leongamornlert D, Saunders E, Dadaev T, Tymrakiewicz M, Goh C, Jugurnauth-Little S, Kozarewa I, Fenwick K, Assiotis I, Barrowdale D, et al. (2014) Frequent germline deleterious mutations in DNA repair genes in familial prostate cancer cases are associated with advanced disease. Br J Cancer 110, 1663-1672.

41. Castro E, Goh C, Olmos D, Saunders E, Leongamornlert D, Tymrakiewicz M, Mahmud N, Dadaev T, Govindasami K, Guy M, et al. (2013) Germline BRCA mutations are associated with higher risk of nodal involvement, distant metastasis, and poor survival outcomes in prostate cancer. J Clin Oncol 31, 1748-1757.

21 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

42. Gallagher DJ, Gaudet MM, Pal P, Kirchhoff T, Balistreri L, Vora K, Bhatia J, Stadler Z, Fine SW, Reuter V, et al. (2010) Germline BRCA mutations denote a clinicopathologic subset of prostate cancer. Clin Cancer Res 16, 2115-2121.

43. Cheng HH, Pritchard CC, Boyd T, Nelson PS & Montgomery B (2016) Biallelic Inactivation of BRCA2 in Platinum-sensitive Metastatic Castration-resistant Prostate Cancer. Eur Urol 69, 992-995.

44. Mateo J, Carreira S, Sandhu S, Miranda S, Mossop H, Perez-Lopez R, Nava Rodrigues D, Robinson D, Omlin A, Tunariu N, et al. (2015) DNA-Repair Defects and Olaparib in Metastatic Prostate Cancer. N Engl J Med 373, 1697-1708.

45. Su X, Shi Y, Zou X, Lu Z-N, Xie G, Yang JYH, Wu C-C, Cui X-F, He K-Y, Luo Q, et al. (2017) Single-cell RNA-Seq analysis reveals dynamic trajectories during mouse liver development. BMC Genomics 18, 946.

46. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Garimella K, Altshuler D, Gabriel S, Daly M, et al. (2010) The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res 20, 1297-1303.

47. Wang K, Li M & Hakonarson H (2010) ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 38, e164.

48. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM & Sirotkin K (2001) dbSNP: the NCBI database of genetic variation. Nucleic Acids Res 29, 308-311.

49. Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL, McCarthy S, McVean GA & Abecasis GR (2015) A global reference for human genetic variation. Nature 526, 68-74.

50. Blokzijl F, Janssen R, van Boxtel R & Cuppen E (2018) MutationalPatterns:

22 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

comprehensive genome-wide analysis of mutational processes. Genome Med 10, 33.

51. Kiselev VY, Kirschner K, Schaub MT, Andrews T, Yiu A, Chandra T, Natarajan KN, Reik W, Barahona M, Green AR, et al. (2017) SC3: consensus clustering of single-cell RNA-seq data. Nat Methods 14, 483-486.

52. Zhang HM, Liu T, Liu CJ, Song S, Zhang X, Liu W, Jia H, Xue Y & Guo AY (2015) AnimalTFDB 2.0: a resource for expression, prediction and functional study of animal transcription factors. Nucleic Acids Res 43, D76-81.

23 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure Legend

Fig. 1. Histological and immunohistochemical staining of the prostate BCC. H&E

staining and the expressions of CK34βE12, P63, PAP and PSA are shown.

Fig. 2. Whole exome sequencing reveals the mutational landscape of prostate BCC.

(A) General description of the exome sequencing of mixtures from genomic

amplification products of prostate BCC single cells and numbers of genetic variations.

(B) Mutational spectra of the prostate BCC based on the variants after SNP filter. (C)

Contributions of COSMIC signatures to the prostate BCC mutational spectra. (D)

Distribution of the variant allele frequency (VAF) of the exonic mutations of prostate

BCC. (E) Summary of the driver (overlapped with The Cancer Gene Census) and

putative sub-clonal (VAF lower than 0.35) mutations of prostate BCC. (F) Percent of

samples mutated in the cBioPortal prostate cancer datasets with those above 1%

shown for the exonic mutated genes in prostate BCC. The red stars indicate Census

driver genes. (G) The items enriched with the exonic mutated genes in prostate BCC

via GSEA/MSigDB canonical pathway enrichment analysis.

Fig. 3. Single-cell RNA-Seq reveals the constituent cell types in the prostate BCC. (A)

Experimental workflow. (B) Heat-map showing the single-cell gene-expression

patterns of the three clusters identified. (C) Heat-map showing the Pearson correlation

coefficients between single-cells. (D) The top 10 items enriched with the gene sets

specifically expressed in each cell cluster via GSEA/MSigDB canonical pathway

24 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

enrichment analysis.

Fig. 4. Single-cell RNA-Seq reveals the transcriptomic features of prostate BCC. (A)

Heat-map showing the expression patterns of top 20 genes specifically and highly

expressed in each cluster, and t-SNE plot of the three clusters. (B) The expression of

selected markers with the median values and the first and third quartiles shown. (C)

Immunohistochemical staining of the specimen with CK14 antibody. (D) TFs

covariance networks of the single-cell data. Each node represents a TF, and each edge

represents correlation between two TFs. The TFs specifically expressed in each

cluster are colored differently.

Fig. 5. Single-cell analysis of prostate BCC provide new insights to prostate cancer.

(A) (left) Heat-map showing the clustering of FACS sorted luminal (CD49f Lo) and

basal (CD49f Hi) cell populations from human benign and cancerous prostates based

on the genes highly expressed in prostate BCC tumor cells; (right) Heat-map showing

the Pearson correlation coefficients between the genes. (Data from Smith et al. PNAS

2015; 112: E6544-E6552). (B) Analysis of prostate cancer samples in similar

approach as (A). (Data from Robinson et al. Cell 2015; 161: 1215–1228). (C)

Comparison of the gene expression profiles of single cells from Cluster 1~3 of

prostate BCC with human prostate CTCs (CTC-candidate and CTC-confirmed), bulk

primary tumors, single cells from prostate cancer cell lines (Pca cell line) and white

blood cells (Data from Miyamoto et al. Science 2015; 349: 1351-1356).

25 bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Fig 1

H&E

CK34βE12

P63

PAP

PSA

100× 200× bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Fig 2 A B C

0.3 1 Whole genome amplification of BCC single cells 0.8 Signature 4

Whole exome sequencing (Mixture of 49 cells) 0.2 Signature 5 0.6 Signature 6 Signature 12 7395 variations after SNP filter Signature 17 Proportion 0.1 Proportion 0.4 Signature 23 91 exonic variations 0.2 Signature 24

0

0 5 genes overlap with Census T>A C>T T>C C>A T>G C>G D E

Group Gene #Chr Position Ref Alt AA change VAF CASC5 15 40903002 A G S86G 0.68 Driver

30 NUTM1 15 34640367 G C V90L 0.63 PTPRC 1 198713275 T A H769Q 0.59 KMT2C 7 151845915 A T I4366N 0.54

20 TBX3 12 115112422 C G A420P 0.42 KRT18 12 53343099 G A V48M 0.33

TNRC18 7 5360200 G A P2200L 0.33 Number of Variant of Number Sub 10 KRT18 12 53343132 G T G59W 0.33 SCUBE3 6 35213742 C T T874I 0.30 - clonal KRT18 12 53343148 C G A64G 0.27

0 ART5 11 3661519 T G E47A 0.25 MST1L 1 17084321 A G C566R 0.17 0.0 0.2 0.4 0.6 0.8 1.0 CCDC40 17 78064020 C A T972K 0.12 Variant Allele Frequency C7orf26 7 6647724 G C A428P 0.09

F G

GO_single_organism_cellular_loca… GO_cell_cycle_process kegg_cell_adhesion_molecules_ca… GO_negative_regulation_of_cataly… GO_regulation_of_hydrolase_activity GO_intracellular_signal_transduction GO_regulation_of_microtubule_ba… GO_cell_cycle GO_microtubule_bundle_formation GO_microtubule_based_process GO_microtubule_cytoskeleton_org… GO_cytoskeleton_organization GO_regulation_of_cell_cycle

0 2 4 6 8 Percent Samples Altered -log10(p) bioRxiv preprint doi: https://doi.org/10.1101/818260; this version posted October 25, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

A Fig 3

Color Key and Histogram Single cells(SC) Captured SC Library prep scRNA-Seq SC transcriptome

Count 40000

0 0 4 8 12 Value B D

kegg_pathways_in_cancer AQP1NFIBEPCAMCDH1KRT14BAMBIKRT5IGF2 NIPAL3NPIPB3DSC3KRT23ACTG2FGFR2MIAFAM60AMALAT1APPCD24PDE9ADSPTRIM2 PAPSS1MZT2AMLXIPRNPS1MARCKSL1KRT15PYGBNRP2CTBP2H19NPIPB4NPIPB5TUFMPTPRF PUM1GABRPPEG3-AS1CAPN12TM4SF1PADI2SCRG1AKAP9FABP7JUPATF4YAP1TFAP2CDSC2 FBXO32TCF7L1GPATCH8NPIPB11PARD6BRBPMSSNX22ERRFI1MPZL1NAV2ATP5A1USP34CRABP2KRT19 MAB21L3KANK4EHFPEG3SETD5PRKYPER2SERPINA3ITGA6NPIPB13ESRP1KLF5PIK3R1PPM1K FGFR1CRIM1SRPK2VITANKRD36KDSRMT2ARTF1ITGB4PIK3R3PRELPSOX11PRKCIFBL MBNL2COL9A2GRHL1IRF6RSF1VCANDSG2KRT8SCNN1GFLNBCSTF3ISG20L2MYCPNISR kegg_prostate_cancer PHKG1CXADRSAA1RPS21SAA2SMIM4RPL11TTC19ARID1APARD3SMA4HP1BP3PCMTD1CCNT2 WDR6ANXA3IGF1RCMYA5TBX3MGST1NUDT5TSTD1BCL11AGAS5RPL7AEFNA1SIK1PGAP1 SORBS2TCERG1FZD7NDUFA7MEIS1BOCFXYD3OBP2AOBP2BART3COL9A1SCHIP1SFRP1GTF2IP1 GOLPH3SNHG3SCUBE3MAEACOL27A1LINC01296CCNL2SMAD5FRMD4AC21orf91TGFBR3RARSEGFRKRT18 43353433504334743165SNRNP40ADNPSNHG25CLDN4CDK6FAT1RPAINMTHFD2SHC4WBP1 C3orf70COL14A1BGNDLC1TIMP1ASPNCOL3A1SULF1COL1A1SLIT2COL1A2THY1COL6A343354 COL6A2PRSS23POSTNAEBP1ISLRLUMPTGFRNECM2CYTORTFPIPDGFRBCOL6A1ID3PCOLCE TNFAIP6RAI14TWIST1ADAM19NRP1MXRA8LHFPLAMA4ZEB1COL18A1EEPD1NTMCRISPLD2GNG11IGFBP7 kegg_melanoma ITGA1CNRIP1CTSKCNN3SFRP2MIR4435-2HGMAP1AIGF1NDUFA4L2SRPX2RGS5MMP11COL5A2HTRA3 PLXDC1SPARCL1EPAS1ZFHX4CAMK2N1MXRA5SERPINE1INPP4BCOL12A1DNAJC15APODDCNSFRP4PRRX1 SYNE1-AS1GSTM3TP53INP1FAPOMDNF2SYT11THBS2DAD1S100A3TPPP3FN1ZNF829SERPING1 PLSCR4SPOCK1PLAC9IFITM1PRAF2ANTXR1ETV1SLC30A7SNED1FBLN1TGFBR2SEMA5AVSTM4NID2 HEG1BICC1PKIGLRP1HSD17B11MINCRMCUBSH3KBP1GNG12NOTCH3ERAP2PRRG1ADIRFIFI16 CRIP1PLXND1SEC23APPP6CCTHRC1ADGRF5ALDH1A1NAV1LOC400043VOPP1ARID5BPARP9RARRES2FHL1 FAM20CRAI2HEYLSGCBMORN2SELENOWFBN1CDH6FABP5COL5A1ANTXR2RPP30MRPL2L3MBTL3 Count LBHHMCN1PLPP1COL8A1EEA1EGLN3EPRSWISP1SYNE2CDK14SLC2A13MESDC2CD99TCF4 kegg_adherens_junction DIO2PTP4A3PEAR1LOC101927311HIGD1BALDH1L2TGFB3SGF29OCEL1 PTNLOC100506388FAM200BABCA6UNC5BADAMTSL3MAFG-AS1RUNX1T1CD109LRP10RNF185OR51E1TMEM219LRRC15 0 60000 TMEM35BFRKTMEM120APDGFRAAXLTRILHOPXTMEM204TIMP3FMO3IGFBP4GPR68PLA2G5SCN7A EBF1STEAP1ALDH1A3IL32TIMM22PINK1LRIG2COPZ2DCAF5SQRDLTACC1LEPROTVGLL3THBS4 ARHGAP28MMP2ZBTB39EXT1PTGISGGT5CHST3ABCA9COL10A1 0 CD248DHRS4GGNBP2SMARCA1MAPRE2CCDC80FAT4GPR176SEC22BATL1POMT1ABCC9PDE3AADCY3 and Histogram C3orf80COL11A1TWSG1APPL2TMEM200AGOLIM4MMP16KCNS3OLFM1MDFIC GRPIGFBP3PLEKHO2PDE1ATIMM9SGIP1EML4C7orf26IRF2BP2NDUFA12LRRC32TBX15RAP1GDS1TRAPPC2B SYNJ2BP-COX16C20orf27FLJ20021DEF8SYNJ2BPTMEM248TMF1MIR5047ADAAPOL1SPHK1PATL1CADM1MAP1B kegg_arrhythmogenic_right_ventricular_… C1SST7LEFEMP2OLFML1MRPL14CXCL12KANK2TNRC6BARHGAP29 Color Key TGFBIMPP5MRVI1HINT2YAE1D1PTPN12RITA1ZNF773PELI2PELOCBX6OSBPL9CEP112AP3B1 MFAP4PLAUA1BGPLXNC1EDNRAITGAVPRKCDBPNAP1L3PLAGL1LRCH3SERPINB6ANGPT2PDLIM3WNT5A

PKD2LENG1TNFRSF12ACDK3IDS 4 OAFSNRNP35GJC1CLCN5LAMP5RASSF8ZNF684CHURC1SPATS2LFGD5-AS1COX17BROXSRGAP2DVAMP5 PJA2TFB1MNLGN4YRNF11CTPS2MCFD2ZNF394SREK1IP1ARHGAP32DSTNIPAL2TMEM11SGMS2FBLN5 FAM179BGLIPR2NAA10LOC101928093GLUD1OSMRUSP13TNS1FAM198BSTT3AOTUD6B-AS1TMX1DYNC1LI1DCBLD2 Value CCZ1BNCOR1MDP1CNIH4LINC00989RFTN2OR2A7GPX8MYO1BLXNZNF280DNDUFA6CPEWWC3 SMC1ATDP2FAM208AANGPTL1C15orf39GNA12PON2CCZ1GLT8D2BTN3A2ZFP90FILIP1LFUT8CFHR1 reactome_cell_cell_communication SEC24DERO1BCLEC2BALKBH5FBXL5CDC5LANO1MAN1A1CAPN2CAV2DOCK5SYNGR1XRRA1PARPBP LINC00963CMTR2ARL1KIAA1551JAK1ATRAIDZNF419MRPS11DPYSL3CYR61FIBING3BP2ROR2EHD2 MKLN1-ASFGFR1OP2SLC12A8DOPEY1RRP15ZBTB38PSENENCDC37L1BTN2A1ZEB2INHBAPARP4VDAC3CMAHP 8 EPS8CRY2KATNBL1PMP22DIS3FRMD6FRYACAD11TAF5LFMODFANCGSPG20KDELR3RBAK HDAC8C10orf10ARFGEF1GTF2H5ATF2DLDMRPL15AGPAT5PGM3TSC22D3C8orf44MALSU1TMEM30AGPC1 APOL2SH3GLB1STX7DLGAP1-AS1MIR99AHGMPRIPRAD51BRAB21LOXSPRY4-IT1S100A6PARP14RPEACLY GUCY1B3CYBRD1IFITM3ATL3PHTF1TAF8SPTBN1RAB3GAP2SARNPHLTF-AS1MRPL33MRPS18CESAMFKBP9P1 CALUITGAEMTCH1FKBP9SAMHD1STEAP2ADAMTS4BCAP29LTBP2TMEM212SLC6A6TRAPPC5SDF2L1PROS1 reactome_cell_junction_organization EDEM1MCAMSLC41A2PDCD5BMP8ACALD1EFTUD2LIN37HDGFRP3IGFBP5PLS3MTERF4RRP1BCRTAP 12 CD302RNASELFBXO42NEIL2ACVR1BABHD10NID1UBE2E2ARHGAP1STRAPZC4H2TBRG1AJUBAMRPS15 MIER1PSMA3PHLDB1SWI5GTF2E1TPM1RBM17USP22PARP8LIMS1TIMP2ZNF324AIDAOXCT1 TEX261MCM6DPYDDDR2UBE2D3URODPDZD11NBEAL1PSME1CAV1ARF6GBP1EIF2B3TOR1AIP1 CD63RIN2DSELERVK3-1TRIM8WHAMMHDAC9CDKL1PSMB4CASC4RNF216A2MLEPROTL1ITPRIPL2 ZNF468SCML1ATG10SLC26A2HOOK3RHBDD3SGPP1CFHSRP54TMEM251LOC100131289GSTA4P4HA1FAM110B STX2LOC143666HAAOTCEAL9KCTD15MAP4DOCK1FAM168ACDK5SSPNB4GALT1SMAP1SSTR2RHOBTB3 SNX2CRIP2THOC2CAMK1DHARSSMOC2GUCY1A3ATF6TSEN15CALRZNF333YIF1AFLRT2XAF1 pid_ajdiss_2pathway SNU13USP31EIF5A2HIVEP3FAM111AZNF30TNFSF10SSSCA1TADA2B MNS1MLECPKN2SDHBRRBP1NDUFA13GTF2BTTLL3CCM2TBC1D15PMS2P3CTDSP1DBNDD2LYPLAL1 KCTD20NREPC1QTNF5ZNF8THRAP3PSMC2STN1RAB40BARL2FRZBDDX39AFAHMYNNLOC105371814 SLC5A3RAB5BPSMD2ATP5C1KRBOX4 CMTM3SQSTM1ASH2LSAFBCSTF1MAPK1IP1LSMIM10L1SPIN1RWDD3LAMP2POGLUT1ZNF484NR1D2SNRPN SPON2LAMTOR3SARSPDE5APTGR1FAM199XPTENMED12CDC42BPANDUFB10PLCE1ZNF451ERHUQCR11 NT5CCHORDC1MAP4K4CIR1NBL1PPEF1COL4A1SAP30BPNBNIL1R1C1RSELENONKMT2CANXA7 ELK4TUBB6SPCS2SLC7A6FAM114A1HOTAIRALG13BEX3USP3SLC35B4NUB1ELOBDEGS1SDC2 Tumor_C02 TPBGSPOPFBXO7CTCFHNRNPUSGCERDXLINC01420LOC284023ZHX1DDX3YKIAA0040HNRNPH2OMA1 kegg_endometrial_cancer Tumor_C05 HIPK1SBDSP1AMPD2INTUANKRD10ZNF277DIDO1ITGB1RBMS3USP7FAM229BMPC1BANF1TAGLN OGTMAGOHMETAP2STRN3TMLHELOC102724552MYO19RNF111PPP1R12AOSR2ZC3H7AMAGI3PLNEMP2 Tumor_C06 SPC25SUMO2TSC22D1TMEM14AUSP2PIGCSMC6SLC2A3COPS8TOMM7DNAJC13ITGB1BP1FRY-AS1C4orf3 Tumor_C07 CD160ARHGAP5TMEM59HSD17B10SPARCCOMTC3orf38SPCS3HNRNPA0ARHGAP17ZNF652MBNL1POLLTCEAL4 EIF4EGUSBSNORA104SUMO3PSMA1ODF2NPTN-IT1THAP4DNAJB11PHAXDIAPH2CNEP1R1HSPBAP1ANXA4 Tumor_C08 ABHD18IGDCC4SBNO1TTC37GSNIDH3AITGBL1FASKLF10NAGKDNASE1L1RAB12SENP6KARS Tumor_C14 REEP3SUCLA2GLG1ZNF721LRRC57ITGB5CCDC82GUCD1ELOVL5ATOX1PDS5BPOLR2KPIP4K2CNNMT Tumor_C15 ANAPC10RNF135CCDC51LYARANKHDTX4BRK1SLC35F1ARF5FSTL1PMEPA1MPDU1UFL1JAG1 reactome_cell_cell_junction_organization STAT1PPP4R3BCRIPTSSR4MORC4TLE1DDB2TMBIM6COA1PURACYP51A1NOC3LSRPXRRAGB DPYSL2TRIM59UBE2FVIMACTN1 Tumor_C16 IRF7CDC26NMINAA20SBDSSYNPO2PRMT7CDH11COX4I1BET1LSSBEBLN3PHCFC2DNAJB4 Tumor_C20 LOC100419170EIF2S1NUTF2EIF3EFXYD6UBE4ATRIM22IFNGR1NDUFC2NORADPTBP2GPX4PRKD1FTO LYRM2RELL1ADAM9FAM8A1WTAPMAGOHBMETTL4RFLNBSUGT1IFT27TFAMEIF3IFKBP11METTL17 Tumor_C22 NBPF8ERGIC2RGS16CENPCNUDT10LOC101929340RGCCNUPL2SERP1HUWE1UBE2D4VRK3RBM3MGST3 Tumor_C23 figure margins too large ATP5JTSTA3LSM3ITFG2ANXA5PLIN2C6orf89EIF4G3MBD3EXOSC8TSR2NUTM2A-AS1LEPRCSPP1 RAB22AINTS10CDH13CBWD6MAD2L1BP Tumor_C24 SLC50A1B4GALT3EIF5BTMEM45ATJP1ERLIN2PQLC3TNKSNADKNKAPUBL5HPRT1MED4-AS1VEGFB MEF2CCCDC174CNPTPT1MAGT1SIPA1L1PTK2USP9XHSD17B7HDDC2PFDN4SERF2TSPAN6PWP1 Tumor_C27 TMED4HGSNATFAM13CMRPL51C3orf14C6orf106MESTRB1CC1WDR45MIR3652ATP5F1OST4ARMCX3PDIA3 Tumor_C28 NDUFV3N4BP2L2CAPZBMFN2TNFRSF21GXYLT1ATP6V1C1ITM2CWASHC2AUBE2Q2P1EMC10NUCB2PGBD1EPB41L2 PKMAPCDD1SRP9VPS8MORF4L1RARBCCDC25ZFYVE19FKBP7HPF1HYPKARPC4IFT52UGGT1 Tumor_C32 ZNF138HSPH1LGALSLSNX6TMBIM4TXNRD1KMT5ADYNLL1LAMTOR1PUM3BLOC1S6AHNAKSNAPC1METTL23 Tumor_C33 NFKBIZZNF639TRA2AKIDINS220ETNK1CLCN3PTRFNBR1FCGRTHNRNPRTBC1D24HSPB1FIS1SDF2 GAL3ST4CHCHD2NDUFA3RBM6RNPC3RNF130VHLSLC35F5NDUFA1SNAP23TSPAN31COPS5TBCBCCNI Tumor_C37 KAT6ALSG1ORC4UBXN4CBWD3VPS13BRNASEKTHUMPD1RNF170PINK1-ASHCFC1R1RNF7ITM2BLOC101928137 ELF1ZNF438NBPF15SCARNA18YWHAEMYH9EIF2S3SLC44A1MED4 Tumor_C38 RPS15AP10EXOSC9C12orf10ZNF605PIGVRAB15RDH11NCOA1TMEM126BGINM1ANXA11CAP1COG3STK3 0 2 4 6 8 10 Tumor_C40 CXCL8LAPTM5RGS1SLCO2B1HCLS1C1QACD14MNDAMS4A7RAPGEF2RBPJTSG101OPHN1NDUFAF1 MS4A6AFCGR2BDOCK8PIK3CGTLR4NCF2FCGR3AIGSF6GPR34FCGR2CTYROBPMSR1FCGR2AHLA-DRB5 Tumor_C43 THEMIS2CCL4L1RNASET2IGSF21HBEGFCYBBFTH1P3CCL4L2CCL3CCL4NFATC2C3AR1C5AR1TNFSF13 Tumor_C49 CCL3L1C1orf162GPR65ITGAXRGS2ICAM1SLC7A7PTPRCC1QCPLAURFGL2C1QBALOX5APITGB2 Tumor_C02 Tumor_C50 GIMAP4CCL3L3ARPC1BHAVCR2TREM2LYZLCP1AP1B1MPEG1CD74SRGNIL1BHLA-DQA1CD83 BMP2KMAP3K8ADAP2BCAT1ACSL1IQGAP2MAFPLA2G16CXCL16CTSSARRB2ITGAMFCER1GGPR132 Cluster_1 VASH1FCGR1ALPAR5RNASE6CD68SIGLEC14LGALS9FPR3SGK1CD53 Tumor_C05 Tumor_C51 SPP1PTGS2LY86CXCL3MLKLFCGR3BCD84SCDFCGR1CPIFI30FCGR1BCYBAMGAT5IKZF1 Tumor_C57 GMFGEPB41L3EVI2BCD86FOLR2TBXAS1LAT2PALD1CCL20PLEKCLEC7ACKLFCXCR4TNFAIP3 LY96PECAM1FERMT3CD163NCF4NLRP3CSF1RAPCSTAB1MPP1SLC15A3MRC1TMEM52BGGTA1P Tumor_C06 Tumor_C58 SP140WDR45BMS4A4AADAM28JAMLLINC01272IL6RCSD1NMBS100A9GPR183CXCL2ALCAMLILRB2 Tumor_C60 MAN2B1DPEP2RNVU1-20LRRC25FYBRAB10CD93HK3RNF149CARD16SIGLEC9LMO2FGD2PILRA Tumor_C62 RAB20CPVLPREX1SLAMF8VAV1KCTD12TMEM192BCL2L1STAT5ACCR1BLNKCIDEBMMDTTC38 Tumor_C07 HCAR3PRDM1EMBLACTBCD4OLR1CCL8HCSTPLCG2RASSF5VMO1DOCK2HCKPYCARD-AS1 Tumor_C63 SAMD9LTNFRSF1BTICAM1LOC105371414CPED1GSKIPSTK40TREM1CAPGTINF2DOK3KLF2CD37PYCARD Tumor_C66 C3LOC102724297DAB2RAC2OTUD1PLA2G7TPP1SLC25A24SLC1A3CSF3RSELPLGRIPK2SLC29A1UCP2 Cluster_2 Tumor_C08 DERAIFI27L2HMGA1JUNBRGS10COTL1GNA13GAPTCSRNP1RASSF4CSF2RBSTARD3NLITGB3ELL2 Tumor_C68 STARD5SLC37A2CASP1KLF4GNAI2NFKBIANXT2CCL2MFSD1MFSD2AAPOC2DAPP1CMKLR1CREBL2 naba_matrisome LCP2MAFBMKL1APH1BCYB5R4FAM49ARRM2BGDAP2FNIP2 Tumor_C70 SNORA75TPD52L2ARHGAP18GKHEXBHPGDSMYO9BARL6IP4CCDC88APHLDA1JAZF1ZMIZ1LINC01094PLTP Tumor_C14 Tumor_C71 C2orf69TRAPPC2LCST3ENGFAM129AAPOC1DNAJC5CMTM7RELLIPADHODHSNCATFE3RHBDF2 SDSARL4CPLBD1SEMA4DNME6CAB39PTPREDIAPH2-AS1CECR1ZNF571SNX10NCOA4SLAEMP3 Tumor_C72 ANKFY1RHOGABCF2MAP3K2SRGAP2BSERPINB9SLC35F6CARD8APBB1IPPLEKHJ1GBP2ETS2FCHO2PA2G4P4 Tumor_C15 Tumor_C73 GPR137BMESDC1GPX1ZDHHC2XYLT1OLFML3GRASPBLOC1S2FPR1CYFIP1MAPKAPK3PDE4BSLC31A2SYK CASP10GPKOWNDUFB2-AS1LOC100130357KANTRFTH1SLC7A5CLDN1NEK6JMJD1CLOC100996842REEP5INIPSDCBP Tumor_C74 SLC16A1PSME2ACRBPRAB26SERPINA1ST3GAL6TMEM220LILRB5SLC47A1SUSD1LINC01125GLDNGPN2CD28 Cluster_3 Tumor_C16 ADPRHVAMP8TNFSF13BEPS15PDGFB naba_core_matrisome Tumor_C76 RNASE1NR4A2CYTH1RAP2BPIP4K2ACEP350ZBTB20-AS4FAM102BGNPDA1 CNPY3HCAR2EBAG9IFI44LNAMPTPLIN3MCL1CTNSID2LBRSYNRGDUSP10PRPS1SH2B3 Tumor_C77 ZNF804ASNHG15PREPTARBP2LOC105370526MFSD5GAS7ALDH3B2SFT2D1CEP170TMCO3SEL1L3CD40MIR155HG Tumor_C20 SRSF5PSAPCCDC112FLIILGMNZNF331HERPUD1ZYXLOC145474 Tumor_C79 LOC100506606DOCK7MTPNPBX3BIRC3FAM175BGPNMBRASSF3GATSAKR1B1TOX4GNG5PLSCR1HES6 Tumor_C81 XXYLT1MIEF2PPCSGBP3NCBP1HSPA7PDP1INSIG1MRPL28S1PR2TMEM206ALOX5GLULKIF3B ATP6V1FC9orf72YWHAHPTPN23RBM47SNORA16ACRY1WFDC21PHMGB2GYPCUSF2ANGLPAR2NDST2 Tumor_C82 NPC2WIPF1UBAC1JMJD6PLEKHB2LYNCYLDSWT1COX14MFHAS1SUGP1HNRNPDLVSIG4NFE2L2MX1 Tumor_C84 CPSF2DENND2DTMED5EGR2CYP2S1CREMADGRE2PEA15DAZAP2CTSDCLEC4AP2RY12MALT1COMMD9 naba_ecm_glycoproteins AP2S1DEF6RASA2AKAP13SCO2ATP6V0BITPK1MAFFYBX1SIRPAKDM2ARBMX2ATP8B4IFI27 Tumor_C85 SERPINB1EIF2B4GOLGA3GRNTADA2AAPRTTAF1AOGFRZNF281MGST2DUSP1HIF1A-AS2PIK3AP1C15orf48 Tumor_C86 ATP13A3MTMR1PPP2R5ESWAP70MX2GAPLINCGSTO1HIPK3MRPL49ATF3MAP1LC3B2ZFP36L2MCM3APRASGEF1B MED29ASLRTFDC1SAT1CD81ST8SIA4ANKRD44CTSBHEXAP2RY13TAF9BRECQLSEL1LTPM3 Color Key Tumor_C87 BORCS8SLC48A1SLC7A8PIGFABRACLATP6V1B2SH3D19IL18IFIH1SKAP2RHOQARPC5LMRPL18MAP1LC3B Tumor_C88 HM13SAR1BFEZ2MR1ACTR3UBXN2BACAA2DUSP2SMPD1UBE2D1ENTPD1SLC30A4CD69NRIP1 Tumor_C89 MYCT1C17orf58CD33LITAFRASSF2CHCHD10RNF144BSCNN1BPTGS1COX5AMGC16275LOC100129034SNORA100FAM188A ARHGAP15BLVRBMIR4647ARPC3LRRC40SATB1TSPAN33SPIDRTGOLN2TXNDC11SETD3UTP14ASHQ1DYNLRB1 pid_integrin1_pathway YIPF5TGFBR1CEP63HMGN4 Tumor_C90 RNF216P1SP110IER5GNB1HNRNPCTGS1TNFAIP2HSP90AA1FAM120AMANBALSM5GALTLOC100288798POLD4 Tumor_C92 TCP1GRWD1S100A5L3MBTL4-AS1ABT1AMZ2ASNSD1PPM1GSUPT5HSDC3SCRN1PTGER4SSR1FUOM and Histogram WASH3PISCUSNORD58CNQO2SAP18STYXL1C6orf62TMEM14BSLC39A4SLC31A1NINJ2STK17BTSPAN14ZNF718 Tumor_C93 TMEM184BMED1CRYL1PXDC1WASHC4XRN2SNX17ZEB2-AS1PLP2SP100SLC40A1IRAK4FUSURB1-AS1 Tumor_C94 RPS19BP1THTPADNTTIP2AMDHD2ARID4AIRF5CDC73NGDNBCAS2BAZ1AMRPS18ABIDSNORA55RAD23A

Tumor_C95 RPP14

Tumor_C05 Tumor_C06 Tumor_C07 Tumor_C08 Tumor_C14 Tumor_C15 Tumor_C16 Tumor_C20 Tumor_C22 Tumor_C23 Tumor_C24 Tumor_C27 Tumor_C28 Tumor_C32 Tumor_C33 Tumor_C37 Tumor_C38 Tumor_C40 Tumor_C43 Tumor_C49 Tumor_C50 Tumor_C51 Tumor_C57 Tumor_C58 Tumor_C60 Tumor_C62 Tumor_C63 Tumor_C66 Tumor_C68 Tumor_C70 Tumor_C71 Tumor_C72 Tumor_C73 Tumor_C74 Tumor_C76 Tumor_C77 Tumor_C79 Tumor_C81 Tumor_C82 Tumor_C84 Tumor_C85 Tumor_C86 Tumor_C87 Tumor_C88 Tumor_C89 Tumor_C90 Tumor_C92 Tumor_C93 Tumor_C94 Tumor_C95 Tumor_C01 Tumor_C04 Tumor_C26 Tumor_C41 Tumor_C42 Tumor_C53 Tumor_C69 Tumor_C10 Tumor_C12 Tumor_C19 Tumor_C34 Tumor_C75 Tumor_C91 Tumor_C01 Tumor_C02 kegg_focal_adhesion Tumor_C04 Tumor_C26 Tumor_C41 Tumor_C42 Count 200 Tumor_C53 naba_matrisome_associated Tumor_C69 Tumor_C10 Tumor_C12

Tumor_C19

0

Tumor_C05 Tumor_C06 Tumor_C07 Tumor_C08 Tumor_C14 Tumor_C15 Tumor_C16 Tumor_C20 Tumor_C22 Tumor_C23 Tumor_C24 Tumor_C27 Tumor_C28 Tumor_C32 Tumor_C33 Tumor_C37 Tumor_C38 Tumor_C40 Tumor_C43 Tumor_C49 Tumor_C50 Tumor_C51 Tumor_C57 Tumor_C58 Tumor_C60 Tumor_C62 Tumor_C63 Tumor_C66 Tumor_C68 Tumor_C70 Tumor_C71 Tumor_C72 Tumor_C73 Tumor_C74 Tumor_C76 Tumor_C77 Tumor_C79 Tumor_C81 Tumor_C82 Tumor_C84 Tumor_C85 Tumor_C86 Tumor_C87 Tumor_C88 Tumor_C89 Tumor_C90 Tumor_C92 Tumor_C93 Tumor_C94 Tumor_C95 Tumor_C01 Tumor_C04 Tumor_C26 Tumor_C41 Tumor_C42 Tumor_C53 Tumor_C69 Tumor_C10 Tumor_C12 Tumor_C19 Tumor_C34 Tumor_C75 Tumor_C91 Tumor_C02 Tumor_C34 Tumor_C75 pid_avb3_integrin_pathway Tumor_C91 0.2 0.6 1 ZYX ZNF804A ZNF718 ZNF571 ZNF331 ZNF281 ZMIZ1ZFP36L2 ZEB2-AS1ZDHHC2 ZBTB20-AS4 YWHAHYIPF5YBX1 XYLT1XXYLT1XRN2 WIPF1WFDC21P WDR45B WASHC4 WASH3P VSIG4VMO1VAV1 VASH1VAMP8UTP14A USF2URB1-AS1 UCP2 UBXN2B UBE2D1 UBAC1 TYROBPTXNDC11 TTC38 TSPAN33 TSPAN14TREM2TREM1TRAPPC2LTPP1 TPM3TPD52L2 TOX4TNFSF13B TNFSF13 TNFRSF1B TNFAIP3TNFAIP2TMEM52BTMEM220 TMEM206TMEM192TMEM184BTMEM14BTMED5TMCO3 TLR4 TINF2TICAM1THTPATHEMIS2 TGS1TGOLN2 TGFBR1 TFE3TCP1 TBXAS1TARBP2TAF9BTAF1ATADA2A SYNRG SYK SWT1SWAP70SUSD1 SUPT5H SUGP1STYXL1STK40STK17BSTAT5A STARD5 STARD3NL STAB1ST8SIA4ST3GAL6 SSR1SRSF5SRGN SRGAP2BSPP1 SPIDR SP140SP110SP100SNX17SNX10 SNORD58C SNORA75 SNORA55 SNORA16A SNORA100 SNHG15 SNCA SMPD1SLCO2B1 SLC7A8 SLC7A7 SLC7A5 SLC48A1 SLC47A1 SLC40A1 SLC39A4 SLC37A2 SLC35F6 SLC31A2 SLC31A1 SLC30A4 SLC29A1 SLC25A24 SLC1A3 SLC16A1 SLC15A3 SLAMF8SLASKAP2 SIRPASIGLEC9 SIGLEC14SHQ1 SH3D19 SH2B3SGK1SFT2D1 SETD3SERPINB9SERPINB1SERPINA1SEMA4D SELPLGSEL1L3SEL1LSDS SDCBP SDC3 SCRN1 SCO2 SCNN1B SCD SATB1SAT1SAR1BSAP18SAMD9LS1PR2 S100A9S100A5RTFDC1 RRM2B RPS19BP1 RPP14RNVU1-20 RNF216P1 RNF149 RNF144B RNASET2RNASE6RNASE1RIPK2RHOQ RHOG RHBDF2 RGS2 RGS10 RGS1 REL REEP5RECQL RCSD1 RBMX2RBM47 RASSF5RASSF4RASSF3RASSF2RASGEF1B RASA2RAP2BRAD23A RAC2 RAB26RAB20 RAB10PYCARD-AS1PYCARDPXDC1 PTPRE PTPRCPTPN23PTGS2PTGS1PTGER4 PSME2PSAP PRPS1PREX1PREP PRDM1 PPP2R5EPPM1GPPCSPOLD4 PLTP PLSCR1 PLP2PLIN3PLEKHJ1 PLEKHB2 PLEKPLCG2 PLBD1PLAURPLA2G7 PLA2G16 PIP4K2APILRAPIK3CGPIK3AP1PIGF PHLDA1 PECAM1PEA15PDP1PDGFB PDE4BPBX3 PALD1PA2G4P4P2RY13 P2RY12 OTUD1 OLR1 OLFML3 OGFR NXT2NRIP1 NR4A2 NQO2 NPC2 NME6 NMB NLRP3 NINJ2 NGDN NFKBIA NFE2L2 NFATC2 NEK6NDUFB2-AS1 NDST2 NCOA4 NCF4 NCF2 NCBP1 NAMPT MYO9BMYCT1MX2MX1MTPNMTMR1 MSR1MS4A7MS4A6AMS4A4AMRPS18A MRPL49 MRPL28 MRPL18 MRC1 MR1 MPP1MPEG1MNDA MMD MLKL MKL1MIR4647 MIR155HG MIEF2MGST2MGC16275 MGAT5MFSD5MFSD2A MFSD1MFHAS1 MESDC1MED29 MED1MCM3APMCL1 MAPKAPK3MAP3K8MAP3K2MAP1LC3B2MAP1LC3B MANBA MAN2B1MALT1MAFFMAFBMAF LYZLYNLY96 LY86 LSM5 LRRC40 LRRC25 LPAR5LPAR2LOC145474 LOC105371414 LOC105370526 LOC102724297 LOC100996842 LOC100506606 LOC100288798 LOC100130357 LOC100129034 LMO2 LITAFLIPA LINC01272 LINC01125 LINC01094 LILRB5 LILRB2 LGMN LGALS9LCP2 LCP1 LBR LAT2LAPTM5LACTB L3MBTL4-AS1KLF4KLF2 KIF3BKDM2A KCTD12 KANTRJUNB JMJD6 JMJD1C JAZF1 JAMLITPK1ITGB3ITGB2ITGAXITGAMISCU IRF5IRAK4IQGAP2INSIG1INIPIL6 IL1B IL18 IKZF1IGSF6IGSF21 IFIH1IFI44LIFI30IFI27L2IFI27 IER5ID2ICAM1HSPA7HSP90AA1HPGDS HNRNPDL HNRNPC HMGN4 HMGB2HMGA1 HM13 HLA-DRB5 HLA-DQA1 HK3 HIPK3 HIF1A-AS2HEXBHEXAHES6HERPUD1 HCST HCLS1 HCK HCAR3 HCAR2 HBEGF HAVCR2 GYPCGSTO1GSKIPGRWD1 GRN GRASPGPX1GPR65 GPR34 GPR183 GPR137B GPR132 GPNMBGPN2 GPKOWGOLGA3 GNPDA1 GNG5 GNB1 GNAI2 GNA13 GMFGGLUL GLDN GK GIMAP4GGTA1P GDAP2GBP3GBP2 GATSGAS7GAPTGAPLINCGALT FYBFUS FUOM FTH1P3 FTH1 FPR3 FPR1FOLR2 FNIP2FLIIFGL2 FGD2 FEZ2FERMT3FCHO2 FCGR3B FCGR3A FCGR2C FCGR2B FCGR2A FCGR1CP FCGR1B FCGR1A FCER1G FAM49AFAM188A FAM175B FAM129A FAM120A FAM102B EVI2BETS2 EPS15EPB41L3ENTPD1 ENG EMP3 EMBELL2 EIF2B4EGR2 EBAG9DYNLRB1 DUSP2 DUSP10 DUSP1 DPEP2DOK3 DOCK8 DOCK7 DOCK2 DNTTIP2DNAJC5 DIAPH2-AS1DHODH DERA DENND2D DEF6 DAZAP2DAPP1DAB2CYTH1 CYP2S1CYLD CYFIP1CYBBCYBACYB5R4 CXCR4 CXCL8 CXCL3 CXCL2 CXCL16 CTSS CTSDCTSBCTNS CST3CSRNP1 CSF3R CSF2RB CSF1R CRYL1 CRY1 CREM CREBL2 CPVLCPSF2CPED1 COX5A COX14 COTL1 COMMD9 CNPY3 CMTM7 CMKLR1 CLEC7A CLEC4A CLDN1 CKLF CIDEB CHCHD10 CEP63CEP350 CEP170 CECR1 CDC73 CD93 CD86 CD84 CD83 CD81 CD74 CD69 CD68 CD53 CD40 CD4 CD37 CD33 CD28 CD163 CD14 CCR1 CCL8 CCL4L2 CCL4L1 CCL4 CCL3L3 CCL3L1 CCL3 CCL20 CCL2 CCDC88A CCDC112 CASP10CASP1 CARD8 CARD16 CAPGCAB39C9orf72 C6orf62 C5AR1 C3AR1 C3 C2orf69 C1QC C1QB C1QA C1orf162 C17orf58 C15orf48 BORCS8 BMP2KBLVRBBLOC1S2 BLNK BIRC3 BIDBCL2L1 BCAT1BCAS2BAZ1A ATP8B4ATP6V1FATP6V1B2ATP6V0BATP13A3 ATF3ASNSD1ASL ARRB2 ARPC5L ARPC3 ARPC1B ARL6IP4ARL4C ARID4A ARHGAP18 ARHGAP15 APRTAPOC2APOC1APH1B APC APBB1IPAP2S1AP1B1ANKRD44 ANKFY1ANG AMZ2AMDHD2 ALOX5AP ALOX5ALDH3B2 ALCAMAKR1B1AKAP13ADPRH ADGRE2 ADAP2ADAM28ACTR3 ACSL1 ACRBP ACAA2ABT1ABRACLABCF2 ZNF829 ZNF8 ZNF773 ZNF721 ZNF684 ZNF652 ZNF639 ZNF605 ZNF484 ZNF468 ZNF451 ZNF438 ZNF419 ZNF394 ZNF333 ZNF324 ZNF30 ZNF280D ZNF277 ZNF138 ZHX1 ZFYVE19ZFP90ZFHX4ZEB2ZEB1ZC4H2 ZC3H7A ZBTB39ZBTB38YWHAEYIF1A YAE1D1XRRA1 XAF1WWC3 WTAP WNT5A WISP1WHAMM WDR45 WASHC2A VSTM4VRK3VPS8 VPS13BVOPP1VIM VHL VGLL3 VEGFBVDAC3 VAMP5USP9XUSP7USP31USP3USP22 USP2USP13UROD UQCR11 UNC5B UGGT1 UFL1 UBXN4 UBL5 UBE4AUBE2Q2P1 UBE2FUBE2E2UBE2D4 UBE2D3 TXNRD1 TWSG1TWIST1TUBB6TTLL3 TTC37 TSTA3TSR2TSPAN6TSPAN31TSG101 TSEN15TSC22D3 TSC22D1 TRIM8TRIM59TRIM22 TRILTRAPPC5TRAPPC2BTRA2A TPT1 TPPP3TPM1TPBGTP53INP1TOR1AIP1 TOMM7TNS1TNRC6B TNKSTNFSF10 TNFRSF21 TNFRSF12A TNFAIP6TMX1TMLHE TMF1 TMEM59TMEM45ATMEM35BTMEM30ATMEM251 TMEM248TMEM219TMEM212TMEM204TMEM200A TMEM14ATMEM126BTMEM120ATMEM11TMED4TMBIM6TMBIM4TLE1TJP1 TIMP3TIMP2 TIMP1TIMM9TIMM22THY1THUMPD1 THRAP3 THOC2 THBS4THBS2THAP4 TGFBR2 TGFBITGFB3TFPITFB1MTFAM TEX261TDP2TCF4 TCEAL9TCEAL4 TBX15TBRG1 TBCBTBC1D24 TBC1D15 TAGLNTAF8T Value C reactome_developmental_biology reactome_axon_guidance

Tumor_C02 Tumor_C05 Tumor_C06 Tumor_C07 Tumor_C08 Tumor_C14 Tumor_C15 0 10 20 30 40 50 Tumor_C16 Tumor_C20 Tumor_C22 Tumor_C23 Tumor_C24 Tumor_C27 Tumor_C28 Tumor_C32 Count Tumor_C33 0 200 Tumor_C37

Tumor_C38 and Histogram reactome_immune_system Tumor_C40 0.2 Tumor_C43 Color Key Tumor_C49 Tumor_C50

Value reactome_adaptive_immune_system Tumor_C51 Tumor_C57 Tumor_C58 0.6 Tumor_C60 Tumor_C62 reactome_hemostasis Tumor_C63 Tumor_C66 Tumor_C68 Tumor_C70 1 kegg_chemokine_signaling_pathway Tumor_C71 Tumor_C72 Tumor_C73 Tumor_C74 Tumor_C02 Tumor_C76 reactome_platelet_activation_signaling_a… Tumor_C05 Tumor_C77 Tumor_C06 Tumor_C79 Tumor_C07 Tumor_C08 Tumor_C81 Tumor_C14 Tumor_C82 kegg_leishmania_infection Tumor_C15 Tumor_C84 Tumor_C16 Tumor_C20 Tumor_C85 Tumor_C22 Tumor_C86 Tumor_C23 Tumor_C87 Tumor_C24 Tumor_C88 reactome_innate_immune_system Tumor_C27 Tumor_C28 Tumor_C89 Tumor_C32 Tumor_C90 Tumor_C33 Tumor_C92 Tumor_C37 Tumor_C93 Cluster_1 reactome_cytokine_signaling_in_immune… Tumor_C38 Tumor_C40 Tumor_C94 Tumor_C43 Tumor_C95 Tumor_C49 Tumor_C01 Tumor_C50 Tumor_C04 Tumor_C51 reactome_g_alpha_i_signalling_events Tumor_C57 Tumor_C26 Cluster_2 Tumor_C58 Tumor_C41 Tumor_C60 Tumor_C42 Tumor_C62 Tumor_C63 Tumor_C53 Tumor_C66 Tumor_C69 reactome_class_a1_rhodopsin_like_recep… Tumor_C68 Tumor_C10 Cluster_3 Tumor_C70 Tumor_C12 Tumor_C71 Tumor_C72 Tumor_C19 Tumor_C73 Tumor_C34 Tumor_C74 Tumor_C75 Tumor_C76 Tumor_C91 Tumor_C77 Tumor_C79 Tumor_C81 Tumor_C82 0 10 20 30 40 50 Tumor_C84 Tumor_C85 Tumor_C86 Tumor_C87 Tumor_C88 Tumor_C89

Tumor_C90

Tumor_C05 Tumor_C06 Tumor_C07 Tumor_C08 Tumor_C14 Tumor_C15 Tumor_C16 Tumor_C20 Tumor_C22 Tumor_C23 Tumor_C24 Tumor_C27 Tumor_C28 Tumor_C32 Tumor_C33 Tumor_C37 Tumor_C38 Tumor_C40 Tumor_C43 Tumor_C49 Tumor_C50 Tumor_C51 Tumor_C57 Tumor_C58 Tumor_C60 Tumor_C62 Tumor_C63 Tumor_C66 Tumor_C68 Tumor_C70 Tumor_C71 Tumor_C72 Tumor_C73 Tumor_C74 Tumor_C76 Tumor_C77 Tumor_C79 Tumor_C81 Tumor_C82 Tumor_C84 Tumor_C85 Tumor_C86 Tumor_C87 Tumor_C88 Tumor_C89 Tumor_C90 Tumor_C92 Tumor_C93 Tumor_C94 Tumor_C95 Tumor_C01 Tumor_C04 Tumor_C26 Tumor_C41 Tumor_C42 Tumor_C53 Tumor_C69 Tumor_C10 Tumor_C12 Tumor_C19 Tumor_C34 Tumor_C75 Tumor_C91 Tumor_C02 Tumor_C92 Tumor_C93 Tumor_C94 Tumor_C95 Tumor_C01 Tumor_C04 -log (p) Tumor_C26 10 Tumor_C41 Tumor_C42 Tumor_C53 Tumor_C69 Tumor_C10 Tumor_C12 Tumor_C19 Tumor_C34 Tumor_C75 Tumor_C91 Tumor_C91 Tumor_C75 Tumor_C34 Tumor_C19 Tumor_C12 Tumor_C10 Tumor_C69 Tumor_C53 Tumor_C42 Tumor_C41 Tumor_C26 Tumor_C04 Tumor_C01 Tumor_C95 Tumor_C94 Tumor_C93 Tumor_C92 Tumor_C90 Tumor_C89 Tumor_C88 Tumor_C87 Tumor_C86 Tumor_C85 Tumor_C84 Tumor_C82 Tumor_C81 Tumor_C79 Tumor_C77 Tumor_C76 Tumor_C74 Tumor_C73 Tumor_C72 Tumor_C71 Tumor_C70 Tumor_C68 Tumor_C66 Tumor_C63 Tumor_C62 Tumor_C60 Tumor_C58 Tumor_C57 Tumor_C51 Tumor_C50 Tumor_C49 Tumor_C43 Tumor_C40 Tumor_C38 Tumor_C37 Tumor_C33 Tumor_C32 Tumor_C28 Tumor_C27 Tumor_C24 Tumor_C23 Tumor_C22 Tumor_C20 Tumor_C16 Tumor_C15 Tumor_C14 Tumor_C08 Tumor_C07 Tumor_C06 Tumor_C05 Tumor_C02 Count 0 1000 2000 0 2 and Histogram Color Key 4 Value 6 8 10 12 A C D B

Tumor_C02

Tumor_C05 CK14 E x p re s s io n (lo g ) E x p re s s io n (lo g ) E x p re s s io n (lo g ) E x p re s s io n (lo g ) Tumor_C06 2 2 2 2 Tumor_C07 1 1 1 1 0 2 4 6 8 0 5 0 2 4 6 8 0 2 4 6 8 0 0 5 0 Tumor_C08 Fig 4

Tumor_C14 bioRxiv preprint Tumor_C15 Tumor_C16 Tumor_C20

COL1A2 THY1 EPCAM CDH1 Tumor_C22 CD14 CD14 CXCL8 Tumor_C23 KRT5 KRT5 KRT14 Tumor_C24 Tumor_C27 Tumor_C28 Tumor_C32 Tumor_C33 Tumor_C37 Tumor_C38

100 Tumor_C40 Tumor_C43 Tumor_C49 Tumor_C50 × Tumor_C51 Tumor_C57 Tumor_C58 E x p re s s io n (lo g 2 ) E x p re s s io n (lo g 2 ) E x p re s s io n (lo g 2 ) E x p re s s io n (lo g 2 ) Tumor_C60 doi: 1 1 1 1 1 1 0 5 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 5 0 0 0 2 Tumor_C62 Tumor_C63 Tumor_C66

Tumor_C68 certified bypeerreview)istheauthor/funder.Allrightsreserved.Noreuseallowedwithoutpermission.

Tumor_C70 https://doi.org/10.1101/818260 Tumor_C71 Tumor_C72 Tumor_C73 Tumor_C74 Tumor_C76 Tumor_C77 Tumor_C79 Tumor_C81 Tumor_C82 Tumor_C84 Tumor_C85 Tumor_C86 Tumor_C87 Tumor_C88

Tumor_C89 figure margins too large E x p re s s io n (lo g ) E x p re s s io n (lo g ) E x p re s s io n (lo g ) E x p re s s io n (lo g ) Tumor_C90 2 2 2 2 Tumor_C92 1 1 1 1 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 0 0 0 Tumor_C93 Tumor_C94

200 Tumor_C95 Tumor_C01

Tumor_C04

Tumor_C26 MNDA × KRT23 CD24 Tumor_C41 BGN Tumor_C42 Tumor_C53 Tumor_C69 Tumor_C10

Tumor_C12

Tumor_C19 Tumor_C34 Tumor_C75 Tumor_C91 PIK3CG TLR4 NCF2 FCGR3A IGSF6 GPR34 FCGR2C TYROBP MSR1 FCGR2A HLA-DRB5 CXCL8 LAPTM5 RGS1 SLCO2B1 HCLS1 C1QA CD14 MNDA MS4A7 ECM2 CYTOR TFPI PDGFRB COL6A1 ID3 PCOLCE C3orf70 COL14A1 BGN DLC1 TIMP1 ASPN COL3A1 SULF1 COL1A1 SLIT2 COL1A2 THY1 COL6A3 DSC3 KRT23 ACTG2 FGFR2 MIA FAM60A MALAT1 APP CD24 PDE9A DSP TRIM2 AQP1 NFIB EPCAM CDH1 KRT14 BAMBI KRT5 IGF2 ;

tsne2 this versionpostedOctober25,2019. -30 30 0 Color Key Color and Histogram and

Cluster_3 Cluster_2 Cluster_1 2000 Tumor_C02 Tumor_C05 Count Tumor_C06 1000 -25

Tumor_C07 0 Tumor_C08 12 10 8 6 4 2 0 Tumor_C14 Tumor_C15 Value Tumor_C16 Tumor_C20 IGF2 tsne1 KRT5 Tumor_C22 BAMBI KRT14 Tumor_C23

CDH1 0 EPCAM NFIB Tumor_C24 AQP1 TRIM2 Tumor_C27 DSP PDE9A Tumor_C28 CD24 APP Tumor_C32 MALAT1 FAM60A MIA Tumor_C33 FGFR2 ACTG2 Tumor_C37

KRT23 25 DSC3 Tumor_C38 COL6A3 THY1 Tumor_C40 COL1A2 SLIT2 Tumor_C43 COL1A1 tsne2 SULF1

COL3A1 Tumor_C49 -30 ASPN TIMP1 Tumor_C50 30 DLC1

BGN 0 COL14A1 Tumor_C51 C3orf70 PCOLCE Tumor_C57 ID3 50 COL6A1 Tumor_C58 PDGFRB TFPI Tumor_C60 CYTOR ECM2 MS4A7 Tumor_C62 MNDA CD14 Tumor_C63 C1QA HCLS1 Tumor_C66 SLCO2B1 RGS1 Tumor_C68 labels LAPTM5 CXCL8 HLA-DRB5 Tumor_C70

FCGR2A -25 MSR1 Tumor_C71 TYROBP FCGR2C Tumor_C72

GPR34 Cluster_3 Cluster_2 Cluster_1 Cluster_3 Cluster_2 Cluster_1

IGSF6 The copyrightholderforthispreprint(whichwasnot FCGR3A Tumor_C73 NCF2 TLR4 Tumor_C74 PIK3CG Tumor_C76 Tumor_C77

Tumor_C79

Tumor_C94 Tumor_C93 Tumor_C92 Tumor_C90 Tumor_C89 Tumor_C88 Tumor_C87 Tumor_C86 Tumor_C85 Tumor_C84 Tumor_C82 Tumor_C81 Tumor_C79 Tumor_C77 Tumor_C76 Tumor_C74 Tumor_C73 Tumor_C72 Tumor_C71 Tumor_C70 Tumor_C68 Tumor_C66 Tumor_C63 Tumor_C62 Tumor_C60 Tumor_C58 Tumor_C57 Tumor_C51 Tumor_C50 Tumor_C49 Tumor_C43 Tumor_C40 Tumor_C38 Tumor_C37 Tumor_C33 Tumor_C32 Tumor_C28 Tumor_C27 Tumor_C24 Tumor_C23 Tumor_C22 Tumor_C20 Tumor_C16 Tumor_C15 Tumor_C14 Tumor_C08 Tumor_C07 Tumor_C06 Tumor_C05 Tumor_C02 Tumor_C91 Tumor_C75 Tumor_C34 Tumor_C19 Tumor_C12 Tumor_C10 Tumor_C69 Tumor_C53 Tumor_C42 Tumor_C41 Tumor_C26 Tumor_C04 Tumor_C01 Tumor_C95

tsne1 Tumor_C81 Tumor_C82 0 Tumor_C84 Tumor_C85 Tumor_C86 Tumor_C87 Tumor_C88 Tumor_C89 Tumor_C90

25 Tumor_C92 Tumor_C93 Tumor_C94 Tumor_C95 Tumor_C01 Tumor_C04 Tumor_C26 Tumor_C41

50 Tumor_C42 Tumor_C53 Tumor_C69 Tumor_C10 Tumor_C12 Tumor_C19 Tumor_C34 Tumor_C75 labels Tumor_C91 Tumor_C20 Tumor_C16 Tumor_C15 Tumor_C14 Tumor_C08 Tumor_C07 Tumor_C06 Tumor_C05 Tumor_C02 Cluster_3 Cluster_2 Cluster_1 Count 0 4000 8000 0 5 and Histogram Color Key Value 10 C B A

15 Count

0 300 Fig 5 0 bioRxiv preprint 2 and Histogram

Color Key Count

Tumor_C10 4 Tumor_C91Tumor_C75 0 40 and Histogram Tumor_C19Tumor_C34 Value Pr9.1.7Pr8.1.2 0

Pr5.1.3 Color Key

Pr5.1.2 6 Pr5.1.1 Tumor_C66Tumor_C12 5

Pr3.1.2 Value HD1.1.w3 Pr15.1.2Pr11.2.3

HD1.1.w2Pr18.1.3 8 Pr23.1.w1 Pr14.3.2 15 Pr2.1.1Pr9.3.4 Pr13.1.1

Pr11.4.3 10 HD1.1.w1Pr12.1.2 Pr23.1.w2Pr16.1.1 Pr9.3.5Pr6.1.4 PriTum7Pr12.1.1 PriTum12PriTum6 BS2771_CancerLo PriTum5 SC_9057 Tumor_C33 SC_9022 PriTum10PriTum3 MO_1071 BS2980_BenignLo PriTum2PriTum4 TP_2034 PriTum8 SC_9062 SC_9001 doi: PriTum11PriTum1 MO_1094 BS2980_CancerLo Tumor_C42PriTum9 X6115237MO_1118 Tumor_C04 MO_1215 Tumor_C53 TP_2061 BS2916_CancerLo Tumor_C01Tumor_C41 X6115242 Tumor_C69 SC_9038 Tumor_C26 MO_1054 BS3337_CancerLo Tumor_C72Tumor_C43 SC_9012 Tumor_C49Tumor_C58 MO_1020 certified bypeerreview)istheauthor/funder.Allrightsreserved.Noreuseallowedwithoutpermission. Tumor_C15 X6115122SC_9031 BS3337_BenignLo https://doi.org/10.1101/818260 Tumor_C08 SC_9007 Tumor_C76Tumor_C79 Tumor_C94 SC_9037TP_2009 BS2506_BenignLo Tumor_C14 SC_9060 Tumor_C68Tumor_C05 SC_9048 Tumor_C37 SC_9017 Tumor_C07 SC_9094 BS2340_BenignLo Tumor_C71Tumor_C23 MO_1179 Tumor_C81Tumor_C82 SC_9034TP_2020 BS2339_BenignLo Tumor_C32Tumor_C28 SC_9071 Tumor_C06 MO_1336 Tumor_C16Tumor_C38 SC_9046 BS2553_CancerLo Tumor_C90 MO_1249 Tumor_C73 SC_9023SC_9043 Tumor_C92Tumor_C50 BS2553_CancerHi Tumor_C89 SC_9049 Tumor_C63 X1115183SC_9061 Tumor_C60Tumor_C02 SC_9086 BS2340_BenignHi Tumor_C85Tumor_C57 MO_1232MO_1221 Tumor_C27Tumor_C77 X6115118 BS2340_CancerHi Tumor_C22Tumor_C95 X1115153X1115202 Tumor_C62 SC_9019 Tumor_C24 MO_1161 Tumor_C51Tumor_C20 MO_1084 BS2506_BenignHi Tumor_C40 MO_1219 Tumor_C84 SC_9008 Tumor_C70Tumor_C88 X6115227 BS2339_CancerHi Tumor_C74Tumor_C93 SC_9092 Tumor_C87 MO_1192 Tumor_C86 X1115157MO_1128 BS2339_BenignHi Pr10.1.3Pr14.1.1 SC_9080 Pr16.1.3 TP_2032 Pr16.1.2Pr14.3.4 SC_9058 BS2980_BenignHi Pr13.1.2Pr10.2.1 X6115234 Pr22.1.12 SC_9081TP_2064 figure margins too large Pr18.1.9 X1115161 BS2980_CancerHi Pr22.1.1Pr10.2.2 SC_9028 Pr1.1.1Pr9.3.3 MO_1339 Pr4.1.1 MO_1337 BS3337_BenignHi Pr9.3.6 X1115156MO_1114 Pr14.3.3Pr14.3.5 Pr11.1.2 MO_1277 BS3337_CancerHi Pr11.1.1 MO_1262SC_9055 Pr18.1.5Pr3.1.1 MO_1316 Pr22.1.7 Pr17.1.1 MO_1184MO_1176 ACTG2 MIA AQP1 IGF2 PDE9A BAMBI MALAT1 DSP APP CD24 CDH1 KRT14 FGFR2 KRT5 DSC3 FAM60A TRIM2 KRT23 NFIB EPCAM Pr16.1.6Pr9.2.2 MO_1202 Pr10.1.2Pr14.2.2 X1115244 Pr11.4.4 SC_9083SC_9093 Pr2.1.2 X6115117 Pr11.2.1Pr11.2.2 TP_2054 DU145.4Pr19.1.3 SC_9047 Pr18.2.4 SC_9091 Pr19.1.5Pr11.4.6 X6115115MO_1244 Pr22.1.10 SC_9030 Pr14.3.1 SC_9059 Pr11.4.2Pr9.1.6 MO_1095 Pr18.1.6Pr19.1.4 X6115121 Pr18.2.3 X6115114SC_9099 Pr22.1.9 SC_9016 Pr22.1.11Pr14.2.4 SC_9072 Pr21.1.3Pr9.1.5 SC_9029 Pr9.3.7 X6115250 Pr21.1.5 SC_9032 Tumor_C02 Pr21.1.6 SC_9097 Lo CD49f Cancer Hi CD49f Cancer Lo CD49f Benign Hi CD49f Benign Pr20.1.2 SC_9073 Tumor_C05 Pr16.1.4Pr15.1.1 X6115233 Tumor_C06 Pr22.1.3 SC_9063 Tumor_C07 Pr6.1.3 MO_1124 Tumor_C08 Pr18.1.2Pr9.2.1 MO_1241 Tumor_C14 ; PC3.5 X6115224 Tumor_C15 PC3.1 X1115154 Tumor_C16

PC3.4 this versionpostedOctober25,2019. PC3.3 SC_9036 Tumor_C20 Pr18.2.1 X6115251 Tumor_C22 Pr19.1.2Pr17.2.2 SC_9026 Pr18.1.1 SC_9018 Tumor_C23 Pr21.1.4 X6115219X6115247 Tumor_C24 LNCaP.2Pr11.3.1 SC_9009 Tumor_C27 DU145.5 TP_2001 Tumor_C28 LNCaP.1 SC_9054 Tumor_C32 LNCaP.R.1LNCaP.R.4 TP_2010 Tumor_C33 LNCaP.4LNCaP.3 X6115123SC_9050 Tumor_C37 LNCaP.5 MO_1014 Tumor_C38 LNCaP.R.2 MO_1040 Tumor_C40 LNCaP.R.3LNCaP.D.3 Tumor_C43 MO_1013

LNCaP.R.5 TP_2060 Tumor_C49 LNCaP.D.1 SC_9010 Tumor_C50 LNCaP.D.5LNCaP.D.2 SC_9068 Tumor_C51 LNCaP.D.4Pr11.4.5 Tumor_C57 Pr21.1.12Pr18.2.2 Tumor_C58

Pr14.2.3 APP MALAT1 CDH1 EPCAM KRT5 KRT14 DSC3 KRT23 FGFR2 ACTG2 TRIM2 BAMBI IGF2 MIA AQP1 PDE9A DSP NFIB FAM60A Tumor_C60 Pr18.1.7 Tumor_C62 Pr14.3.6Pr22.1.2 Tumor_C63 Pr14.2.5Pr14.2.6 Tumor_C66 Pr21.1.11 Tumor_C68 Pr21.1.10Pr21.1.1 Tumor_C70 Pr21.1.9 Tumor_C71 Pr11.4.1 Tumor_C72 Pr21.1.2Pr21.1.7 Tumor_C73 Pr22.1.4Pr9.1.1 Tumor_C74 Pr9.3.2 Tumor_C76 Pr9.1.2 Tumor_C77 DU145.2PC3.2 Tumor_C79 Pr9.1.3 Tumor_C81 Pr9.3.1 Tumor_C82 Pr22.1.5Pr6.1.5 Tumor_C84 Pr20.1.1Pr4.1.2 Tumor_C85 DU145.3 Tumor_C86 DU145.1 Tumor_C87 Pr14.2.1VCaP.3 Tumor_C88 VCaP.2VCaP.5 Tumor_C89 Pr11.3.2 Tumor_C90 VCaP.1 Tumor_C92 Pr21.1.8VCaP.4 Tumor_C93 Pr17.1.2Pr18.1.8 Tumor_C94 Pr22.1.8 Count Tumor_C95 Pr11.3.3Pr17.2.1 0 Tumor_C01 Pr6.1.1 Count Tumor_C04 Count Pr18.1.4 0 Tumor_C26 and Histogram Pr22.1.6Pr9.3.8 0 40 and Histogram Tumor_C41 0 30 Pr6.1.2 Tumor_C42 -1 Pr9.1.4 -1 Color Key Color Key Tumor_C53 Tumor_C69

PIK3CG TLR4 NCF2 FCGR3A IGSF6 GPR34 FCGR2C TYROBP MSR1 FCGR2A HLA-DRB5 CXCL8 LAPTM5 RGS1 SLCO2B1 HCLS1 C1QA CD14 MNDA MS4A7 ECM2 CYTOR TFPI PDGFRB COL6A1 ID3 PCOLCE C3orf70 COL14A1 BGN DLC1 TIMP1 ASPN COL3A1 SULF1 COL1A1 SLIT2 COL1A2 THY1 COL6A3 DSC3 KRT23 ACTG2 FGFR2 MIA FAM60A MALAT1 APP CD24 PDE9A DSP TRIM2 AQP1 NFIB EPCAM CDH1 KRT14 BAMBI KRT5 IGF2 Tumor_C10 Tumor_C12 Value Value

1 Tumor_C19

Tumor_C34 0 0 and Histogram Tumor_C75

Color Key Tumor_C91 Tumor_C15 Tumor_C14 Tumor_C08 Tumor_C07 Tumor_C06 Tumor_C05 Tumor_C02 Value 1 1 2 MALAT1 NFIB PDE9A IGF2 3 NFIB PDE9A Count FAM60A MALAT1 0 DSP ACTG2 TRIM2 CD24 0 EPCAM EPCAM Tumor_C02 Tumor_C06Tumor_C05 BAMBI Tumor_C08Tumor_C07 and Histogram CDH1

Tumor_C14 White blood cell PCa Primary tumor CTC CTC Cluster_3 Cluster_2 Cluster_1 Tumor_C16Tumor_C15 TRIM2 Tumor_C22Tumor_C20 Tumor_C24Tumor_C23 BAMBI Tumor_C27 Tumor_C28 Color Key Tumor_C33Tumor_C32 MIA Tumor_C37 Tumor_C38 MIA The copyrightholderforthispreprint(whichwasnot Tumor_C43Tumor_C40 5 Tumor_C50Tumor_C49 CDH1 Tumor_C51 Tumor_C57 - - APP Tumor_C58 cell line

Tumor_C60 Value Tumor_C62 confirmed candidate KRT23 Tumor_C66Tumor_C63 Tumor_C68 Tumor_C71Tumor_C70 AQP1 Tumor_C73Tumor_C72 KRT14 Tumor_C76Tumor_C74 Tumor_C77 IGF2 Tumor_C81Tumor_C79 Tumor_C82 FAM60A

Tumor_C84 10 Tumor_C85 Tumor_C86 Tumor_C88Tumor_C87 ACTG2 Tumor_C89 KRT5 Tumor_C92Tumor_C90 Tumor_C93 Tumor_C95Tumor_C94 FGFR2 Tumor_C04Tumor_C01 DSC3 Tumor_C41Tumor_C26 Tumor_C42 Tumor_C69Tumor_C53 KRT23 Tumor_C10 AQP1

Tumor_C12 Tumor_C19 Tumor_C75Tumor_C34

Tumor_C91 DSC3 Pr10.1.2 APP Pr10.1.3

Pr11.1.2Pr10.2.1 15 Pr11.2.3Pr11.2.1 KRT14 Pr12.1.2Pr12.1.1 DSP Pr13.1.1 Pr14.3.2Pr13.1.2 Pr15.1.2Pr14.3.3 Pr16.1.1 KRT5 FGFR2 Pr16.1.3Pr16.1.2 Pr16.1.6Pr16.1.4 Pr18.1.3Pr18.1.2 Pr18.1.9

Pr18.2.4 NFIB IGF2 PDE9A MALAT1 ACTG2 CD24 EPCAM BAMBI TRIM2 MIA CDH1 KRT23 KRT14 FAM60A KRT5 DSC3 AQP1 APP DSP FGFR2 Tumor_C10 Pr20.1.2 MALAT1 PDE9A NFIB FAM60A DSP TRIM2 EPCAM CDH1 BAMBI MIA APP AQP1 IGF2 ACTG2 FGFR2 KRT23 DSC3 KRT14 KRT5 Tumor_C75 Pr22.1.3Pr22.1.1 Tumor_C91 Pr9.1.5 Tumor_C19Tumor_C34 Pr9.2.1Pr9.1.7 Pr8.1.2 Pr9.3.2Pr9.2.2 Pr9.1.7 Pr9.3.4Pr9.3.3 Pr5.1.2Pr5.1.3 Pr9.3.5 Pr5.1.1 Pr2.1.1Pr1.1.1 Tumor_C12 Pr3.1.2Pr2.1.2 Tumor_C66Pr3.1.2 Pr4.1.1 HD1.1.w3 Pr5.1.1Pr4.1.2 Pr11.2.3 Pr5.1.2 Pr15.1.2 Pr5.1.3 HD1.1.w2 Pr6.1.4Pr6.1.3 Pr18.1.3 Pr8.1.2 Pr23.1.w1 Pr11.1.1Pr10.2.2 Pr14.3.2Pr9.3.4 Pr11.2.2 Pr2.1.1 Pr11.3.2Pr11.3.1 Pr11.4.3Pr13.1.1 Pr11.4.1Pr11.3.3 HD1.1.w1 Pr11.4.3Pr11.4.2 Pr12.1.2 Pr11.4.5Pr11.4.4 Pr23.1.w2 Pr11.4.6 Pr16.1.1Pr6.1.4 Pr14.2.1Pr14.1.1 Pr9.3.5 Pr14.2.3Pr14.2.2 PriTum7Pr12.1.1 Pr14.2.4 PriTum12 Pr14.2.6Pr14.2.5 PriTum5PriTum6 Pr14.3.4Pr14.3.1 Pr14.3.6Pr14.3.5 Tumor_C33PriTum10 Pr15.1.1 PriTum4PriTum3 Pr17.1.2Pr17.1.1 PriTum2 Pr17.2.2Pr17.2.1 PriTum8 Pr9.1.1 PriTum1 Pr9.1.3Pr9.1.2 PriTum11 Pr9.1.4 PriTum9 Pr9.3.1Pr9.1.6 Tumor_C42 Pr9.3.6 Tumor_C04 Pr9.3.7 Tumor_C53 Pr18.1.1Pr9.3.8 Tumor_C01Tumor_C41 Pr18.1.5Pr18.1.4 Tumor_C69 Pr18.1.6 Tumor_C26 Pr18.1.7 Tumor_C43 Pr18.2.1Pr18.1.8 Tumor_C72 Pr18.2.2 Tumor_C58 Pr19.1.2Pr18.2.3 Tumor_C49 Pr19.1.3 Tumor_C08Tumor_C15 Pr19.1.5Pr19.1.4 Tumor_C79 Pr21.1.1Pr20.1.1 Tumor_C94Tumor_C76 Pr21.1.10 Tumor_C14 Pr21.1.12Pr21.1.11 Tumor_C05 Pr21.1.3Pr21.1.2 Tumor_C68 Pr21.1.5Pr21.1.4 Tumor_C07Tumor_C37 Pr21.1.6 Tumor_C23 Pr21.1.8Pr21.1.7 Tumor_C71 Pr22.1.10Pr21.1.9 Tumor_C81Tumor_C82 Pr22.1.11 Tumor_C28 Pr22.1.12Pr22.1.2 Tumor_C06Tumor_C32 Pr22.1.5Pr22.1.4 Tumor_C38 Pr22.1.7Pr22.1.6 Tumor_C16 Pr22.1.8 Tumor_C90 Pr22.1.9Pr3.1.1 Tumor_C73 Pr6.1.2Pr6.1.1 Tumor_C92Tumor_C50 Pr6.1.5 Tumor_C63Tumor_C89 PriTum2PriTum1 Tumor_C02 PriTum4PriTum3 Tumor_C60 PriTum11PriTum10 Tumor_C57 PriTum12 Tumor_C77Tumor_C85 PriTum6PriTum5 PriTum8PriTum7 Tumor_C95Tumor_C27 PriTum9 Tumor_C62Tumor_C22 DU145.2DU145.1 Tumor_C24 DU145.4DU145.3 Tumor_C20 LNCaP.1DU145.5 Tumor_C40Tumor_C51 LNCaP.2 Tumor_C84 LNCaP.4LNCaP.3 Tumor_C88 LNCaP.D.1LNCaP.5 Tumor_C70 LNCaP.D.2 Tumor_C74Tumor_C93 LNCaP.D.4LNCaP.D.3 Tumor_C86Tumor_C87 LNCaP.R.1LNCaP.D.5 Pr14.1.1 LNCaP.R.3LNCaP.R.2 Pr10.1.3 LNCaP.R.4 Pr16.1.3 LNCaP.R.5PC3.1 Pr14.3.4 PC3.3PC3.2 Pr16.1.2 PC3.4 Pr13.1.2Pr10.2.1 VCaP.1PC3.5 Pr22.1.12Pr18.1.9 VCaP.3VCaP.2 Pr10.2.2 VCaP.5VCaP.4 Pr22.1.1 Pr23.1.w1 Pr9.3.3 Pr23.1.w2HD1.1.w1 Pr4.1.1Pr1.1.1 HD1.1.w2 Pr14.3.5Pr9.3.6 HD1.1.w3 Pr14.3.3 AQP1 NFIB EPCAM CDH1 KRT14 BAMBI KRT5 IGF2 Pr11.1.1Pr11.1.2Pr3.1.1 Pr22.1.7Pr18.1.5 Pr17.1.1Pr9.2.2 Pr14.2.2Pr16.1.6 Pr11.4.4Pr10.1.2Pr2.1.2 Pr11.2.1Pr11.2.2 DU145.4Pr19.1.3 Pr11.4.6Pr18.2.4 Pr22.1.10Pr19.1.5 Pr11.4.2Pr14.3.1Pr9.1.6 Pr18.1.6Pr19.1.4 Pr22.1.9Pr18.2.3 Pr22.1.11Pr14.2.4 Pr21.1.3Pr9.3.7Pr9.1.5 Pr21.1.6Pr21.1.5 Pr15.1.1Pr20.1.2 Pr22.1.3Pr16.1.4 Pr18.1.2Pr6.1.3 Pr9.2.1PC3.1PC3.5 PC3.3PC3.4 Pr17.2.2Pr18.2.1 Pr18.1.1Pr19.1.2 LNCaP.2Pr11.3.1Pr21.1.4 LNCaP.1DU145.5 LNCaP.R.1LNCaP.R.4 LNCaP.4LNCaP.3 LNCaP.R.2LNCaP.5 LNCaP.R.5LNCaP.R.3LNCaP.D.3 LNCaP.D.2LNCaP.D.1 LNCaP.D.4LNCaP.D.5 Pr18.2.2Pr11.4.5 Pr21.1.12Pr18.1.7Pr14.2.3 Pr14.3.6Pr22.1.2 Pr14.2.5Pr14.2.6 Pr21.1.11Pr21.1.1 Pr21.1.10Pr21.1.9 Pr21.1.2Pr21.1.7Pr11.4.1 Pr22.1.4Pr9.1.1 Pr9.1.2Pr9.3.2 DU145.2PC3.2 Pr6.1.5Pr9.3.1Pr9.1.3 Pr22.1.5Pr4.1.2 DU145.3Pr20.1.1 DU145.1VCaP.3 Pr14.2.1VCaP.5 Pr11.3.2VCaP.2VCaP.1 Pr21.1.8VCaP.4 Pr17.1.2Pr18.1.8 Pr17.2.1Pr22.1.8 Pr18.1.4Pr11.3.3Pr6.1.1 Pr22.1.6Pr9.3.8 Pr9.1.4Pr6.1.2 PIK3CG TLR4 NCF2 FCGR3A IGSF6 GPR34 FCGR2C TYROBP MSR1 FCGR2A HLA-DRB5 CXCL8 LAPTM5 RGS1 SLCO2B1 HCLS1 C1QA CD14 MNDA MS4A7 ECM2 CYTOR TFPI PDGFRB COL6A1 ID3 PCOLCE C3orf70 COL14A1 BGN DLC1 TIMP1 ASPN COL3A1 SULF1 COL1A1 SLIT2 COL1A2 THY1 COL6A3 DSC3 KRT23 ACTG2 FGFR2 MIA FAM60A MALAT1 APP CD24 PDE9A DSP TRIM2 AQP1 NFIB EPCAM CDH1 KRT14 BAMBI KRT5 IGF2