- Open Access
An integrative genomics approach for identifying novel functional consequences of PBRM1 truncated mutations in clear cell renal cell carcinoma (ccRCC)
© The Author(s). 2016
- Published: 22 August 2016
The Erratum to this article has been published in BMC Genomics 2016 17:821
Clear cell renal cell carcinoma (ccRCC) is the most common type of kidney cancer. Recent large-scale next-generation sequencing analyses reveal that PBRM1 is the second most frequently mutated gene harboring many truncated mutations and has a suspected tumor suppressor role in ccRCC. However, the biological consequences of PBRM1 somatic mutations (e.g., truncated mutations) that drive tumor progression in ccRCC remain unclear.
In this study, we proposed an integrative genomics approach to explore the functional consequences of PBRM1 truncated mutations in ccRCC by incorporating somatic mutations, mRNA expression, DNA methylation, and microRNA (miRNA) expression profiles from The Cancer Genome Atlas (TCGA). We performed a systematic analysis to detect the differential molecular features in a total of 11 ccRCC samples harboring PBRM1 truncated mutations from the 33 “pan-negative” ccRCC samples. We excluded the samples that had any of the five high-confidence driver genes (VHL, BAP1, SETD2, PTEN and KDM5C) reported in ccRCC to avoid their possible influence in our results.
We identified 613 differentially expressed genes (128 up-regulated and 485 down-regulated genes using cutoff |log2FC| < 1 and p < 0.05) in PBRM1 mutated group versus “pan-negative” group. The gene function enrichment analysis revealed that down-regulated genes were significantly enriched in extracellular matrix organization (adjusted p = 2.05 × 10−7), cell adhesion (adjusted p = 2.85 × 10−7), and ion transport (adjusted p = 9.97 × 10−6). Surprisingly, 26 transcriptional factors (TFs) genes including HOXB9, PAX6 and FOXC1 were found to be significantly differentially expressed (23 over expressed TFs and three lower expressed TFs) in PBRM1 mutated group compared with “pan-negative” group. In addition, we identified 1405 differentially methylated CpG sites (targeting 1308 genes, |log2FC| < 1, p < 0.01) and 185 significantly altered microRNAs (|log2FC| < 1, p < 0.05) associated with truncated PBRM1 mutations. Our integrative analysis suggested that methylation and miRNA alterations were likely the downstream events associated with PBRM1 truncation mutations.
In summary, this study provided some important insights into the understanding of tumorigenesis driven by PBRM1 truncated mutations in ccRCC. The approach may be applied to many driver genes in various cancers.
- Clear cell renal cell carcinoma (ccRCC)
- Driver gene
Renal cell carcinoma (RCC) is the most common type of kidney cancer (>85 %), which causes ~3 % deaths in men in the United States every year [1, 2]. RCC can be classified into four clinical subtypes including clear cell renal cell carcinoma (ccRCC), papillary RCC (pRCC), chromophobe RCC (chRCC), and renal oncocytoma (RO). Among them, ccRCC is the most common type representing 75–85 % of all RCC cases [2, 3]. Unlike other cancer types that are found to have recurrent mutations in oncogenes [4–7], ccRCC tumors are mainly associated with somatic mutations in tumor suppressor genes such as VHL, PBRM1, BAP1 and SETD2 [8–10].
PBRM1 (Polybromo-1, pb1, encoding BAF180 protein), which maps to 3p21, plays an ATP-dependent chromatin-remodeling role as a subunit of the SWI/SNF (SWItch/Sucrose Non-Fermentable) complex [11–13]. PBRM1 is found to mediate gene regulation of cell growth, migration, proliferation and differentiation in multiple cancer types including kidney, bladder, and breast. Among these cancer types, PBRM1 is one of the most frequently mutated and studied genes in ccRCC than any other cancer types [11, 12, 14–18]. In ccRCC, PBRM1 is the second most frequently mutated gene; it is observed in ~40 % of tumor cases and functions as a driver tumor suppressor gene [3, 9, 10, 13, 18–20]. PBRM1 mutations in ccRCC samples may lead to a dysregulation of several critical cell signaling pathways including actin-based motility by rho, tight junction signaling, axonal guidance signaling and germ cell-sertoli cell junction signaling . Furthermore, mutations in PBRM1 are identified as the root of tumor evolution in a subgroup of ccRCC . While previous studies have focused on the exploration of particular downstream genes and pathways directly regulated by PBRM1 gene, an in-depth integrative analysis on the biological consequences of PBRM1 truncated mutations has not been done yet. Such an analysis is important because tumor suppressor genes play function largely through truncated mutations .
Here, we performed an integrative genomics analysis to investigate the biological consequences of truncated PBRM1 mutations in ccRCC. We downloaded multiple -omics data including RNA-Seq, DNA methylation, and microRNA-Seq data of ccRCC samples from The Cancer Genome Atlas (TCGA). We systemically compared molecular features in a total of 11 mutated PBRM1 samples with those in 33 “pan-negative” samples; and those samples were all exclusive of any of the five known ccRCC driver genes (VHL, BAP1, SETD2, PTEN and KDM5C) [13, 15]. The approach allowed us to maximally reduce the noise from the observed molecular signals. We identified a substantial proportion of molecular alterations including changes in gene expression, DNA methylation, and dysregulation of microRNAs (miRNAs) that were significantly associated with truncated PBRM1 mutations, as well as the follow up pathway, co-expression network, and hypothesized mechanism analysis.
Workflow for defining PBRM1-mutated and “pan-negative” sample groups
Identification of transcriptional factors from differentially expressed genes associated with PBRM1 truncated mutations
Widespread epigenetic silencing associated with PBRM1 truncated mutations
miRNA dysregulation associated with PBRM1 truncation mutations
Integrated analysis for PBRM1 truncated mutations in ccRCC
This study highlights the association between PBRM1 truncated mutations and decreased extracellular matrix organization, cell adhesion, ion transport and tissue development. This suggests that PBRM1 plays an important regulatory role in cell-cell crosstalk in the tumorigenesis of ccRCC. In this study, there are more differentially methylated genes (1308 genes) than differential expressed genes (613 genes) in PBRM1 mutated group, suggesting a complicated pre-transcriptional level regulation with DNA methylation involved in PBRM1 mutations.
Studying the downstream events of a driver gene has become important now because the scientific community has witnessed large amount of genomic data allowing the sample stratification by driver mutation and also because a driver gene may lead to many critical biological events linking to tumorigenesis or drug treatment [31, 32]. We recently develop approaches to study the downstream events of a specific mutation in a driver gene (BRAFV600E and NRASQ61) in melanoma [4, 5]. To our knowledge, this is the first study to integrate pre-transcriptional and post-transcriptional level data to investigate the main effects of a driver gene (PBRM1) through its truncated mutations in a cancer (ccRCC). Observations in this study are based on 11 PBRM1 mutated and 33 “pan-negative” ccRCC samples, which may have some bias because of the small sample size. However, by an integrative analysis of multiple -omics data, we could still achieve reliable results for further validation. As we did similarly in melanoma [4, 5], the stratification of samples by driver mutation only (cases) and “pan-negative” samples (controls) would likely increase the power because it effectively removed the noise from similar samples with other driver mutations. This is especially important in cancer genomics studies because driver mutations may affect the same or similar signaling pathways (e.g., Ras pathway). Our results suggest that PBRM1 mutations are an important event in the early stage of ccRCC tumor genetics, which paves the way for further PBRM1-related research in ccRCC. To excluded the influences of other driver genes and highlight the effects of PBRM1 in ccRCC, we defined the “pan-negative” ccRCC sample set by excluding samples that contained somatic mutations in any of the five well-known driver genes in ccRCC. Future validation may apply the similar strategies. Our integrative analysis using methylation, gene expression, and miRNA expression is the first to study the PBRM1 truncation mutation specific dysfunction in co-expressed networks. All mutations in 11 PBRM1 mutated samples are truncation mutations, which signify dysfunction state of PBRM1 as a tumor suppressor gene in ccRCC.
There are several limitations in this study. First, how our results are related to the influence of PBRM1 on tumor prognosis needs further investigation because previous studies suggest the association between PBRM1 mutations and prognosis of ccRCC is still unclear [13, 22, 33, 34]. In addition, copy number variants of PBRM1 are not considered either since we only focus on the downstream consequences that associate with early somatic mutation events in PBRM1. No validation cohorts of PBRM1 have involved in this study yet because of the limited results available related to PBRM1 at the current stage. We hope more reports will become available from other groups in the near future so that our results may be experimentally validated. Our analysis focuses on the gene level changes that associated with PBRM1 truncated mutation, in which protein level changes were not considered because of the complicated regulation from gene expression level to protein level.
PBRM1 is found to be highly mutated in several cancer types. It is most frequently mutated in ccRCC. Loss of function and expression of PBRM1 was less common in non-ccRCC than in ccRCC, suggesting a specific regulatory role of PBRM1 truncation mutations in ccRCC . In breast cancer, PBRM1 is shown to be a core regulator of p21 ; however, we could not find a similar pattern in ccRCC. The result suggests that PBRM1 may act differently through its regulation mechanisms in different cancer types. Future studies to dissect the role of PBRM1 in different cancer types would be helpful to better understand the mechanisms of PBRM1 truncation mutations and tumorigenesis. More cancer genomic data is expected from large consortia like the International Cancer Genome Consortium (ICGC). So, a follow up study is needed in future.
Our study investigated molecular alterations including gene expression, methylation, and miRNA expression that associated with PBRM1 truncation mutations in clear cell renal cell carcinoma. Our analysis results identified 613 differentially expressed genes, 1308 differentially methylated genes and 185 differentially expressed miRNAs between PBRM1 mutated group and “pan-negative” group. Hypothesized mechanisms of PBRM1 mutations in ccRCC were explored based on the integrative analysis results. Our results provide some important insights into the PBRM1 regulation in the tumor development of ccRCC.
Summary of ccRCC samples
A total of 548 ccRCC (KIRC) samples were downloaded from TCGA. Level 2 results from both BI Mutation Calling and BCM Mutation Calling were utilized to find somatic mutations in all samples. 177 of 548 ccRCC samples (32.3 %) were identified to have PBRM1 mutations and 371 samples (67.7 %) were identified as PBRM1 non-mutated or control samples. To eliminate the influence of other driver genes, five well-known mutation genes (VHL, BAP1, SETD2, PTEN and KDM5C) were suggested as highly potential driver genes of ccRCC based on the somatic mutation results and earlier researches . Samples with somatic mutations of those five genes were excluded from both mutated and non-mutated PBRM1 samples, resulting in 31 PBRM1 mutated samples and 109 “pan-negative” samples (Fig. 1a). Finally, 11 PBRM1 mutated samples and 33 “pan-negative” samples that had DNA methylation, gene expression, and miRNA expression data were utilized for all the analyses in this study.
RNA-Seq and miRNA-Seq data pre-processing and differential expression analysis
RNA-Seq and miRNA-Seq data were downloaded from IlluminaHiSeq_RNASeqV2 and BCGSC IlluminaHiSeq_miRNASeq platform in TCGA database, respectively. Level 3 data were utilized to find RNA expression and miRNA expression. In each group, genes/miRNAs with no expression were removed, while only genes/miRNAs with counts per million (cpm) >1 in at least two samples were kept for further analysis. edgeR package  in R software was used in differential RNA-Seq and miRNA expression analysis. We defined significantly DEGs or differentially expressed miRNAs if they had |log2FC| > 1 and p < 0.05. MiRNA target genes were retrieved from databases TarBase  and miRTarBase .
Illumina HumanMethylation450K BeadChip Kit containing 486,428 CpG sites was used to explore DNA methylation profile on the genome scale. Probes targeting the X and Y chromosome, probes containing a single-nucleotide polymorphism (SNP) within five base pairs of CpG site, and probes that had no reference gene location were also removed. In total, 312,777 probes were kept for further analysis. β-values that ranged between 0 and 1 were used to represent the relative methylation level, which was measured as logistic transformation of the ratio of the methylated probe intensity over all methylation probe intensities . β-difference value (differences between β-values) was used to characterize different methylation levels between PBRM1 mutated group and non-mutated “pan-negative” group. All methylation analysis was performed in R/Bioconductor packages . Samr package in R software  was used to calculate the significance of each CpG site. Probes with |β-difference| > 0.15 and p < 0.01 were selected as differentially methylated probes, and the gplots package in R software was used to obtain a heatmap of differentially methylated probes.
Gene function and pathway enrichment analysis
The ClueGO plugin  in Cytoscape software  was used for gene function and pathway enrichment analysis. Catalogues in GO Biological Process, KEGG, REACTOME and WikiPathways databases that catalogued in ClueGo were applied for the functional enrichment analysis. The Benjamin-Hochberg method  was used in the adjustment of p (false discovery rate), and other parameters were retained as default in GlueGO. Gene sets or pathways with adjusted p < 0.05 were retained for further analysis. Transcription factors were annotated based on the TRANSFAC database (downloaded on April 1, 2015) .
PBRM1 mutation specific, differentially regulated co-expression network
The Pearson correlation coefficient in R software was used to calculate the correlation of each pair on all the 14270 genes that were extracted from RNA-Seq results after excluding low expression genes in PBRM1 mutated group. The top 5 % co-expressed gene pairs were kept as co-expressed and protein-protein interactions from PINA2  were used to find out the relationships between co-expressed genes, which resulted in a PBRM1 mutation specific background network that contains 335,726 gene interaction pairs. 128 up-regulated genes and miRNAs with targets in up-regulated genes were mapped into the reference network, resulting in a PBRM1 mutation specific, up-regulated co-expression network. 485 down-regulated genes and miRNAs with targets in down-regulated genes were mapped into the reference network, resulting in a PBRM1 mutation specific, down-regulated co-expression network. To explore the essential genes associated with PBRM1 mutations, only 33 hyper-down genes and their first neighbors were kept, resulting in PBRM1 mutation specific, down-regulated core co-expression network. The Cytoscape software was used to make the network visualization, with genes that have three or more degrees being shown in Fig. 5a and b.
ccRCC, clear-cell renal cell carcinoma; KIRC, kidney renal clear cell carcinoma; TCGA, the cancer genome Atlas.
The authors would like to thank Drs. Wei Jiang, Feixiong Cheng, Junfei Zhao, Peilin Jia and Ramkrishna Mitra for valuable suggestion and discussion on the data analysis. We thank Vanderbilt Advanced Computing Center for Research & Education (ACCRE) for providing computing resources and support.
Publication of this article was charged from the faculty retention funds to Dr. Zhao from Vanderbilt University.
This article has been published as part of BMC Genomics Volume 17 Supplement 7, 2016: Selected articles from the International Conference on Intelligent Biology and Medicine (ICIBM) 2015: genomics. The full contents of the supplement are available online at http://0-bmcgenomics.biomedcentral.com.brum.beds.ac.uk/articles/supplements/volume-17-supplement-7.
This work was partially supported by National Institutes of Health (NIH) grants (R01LM011177 and R21CA196508) and Ingram Professorship Funds (to Z.Z.). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Availability of data and materials
All the data used in this study is from the public sources with the links being included in the publication. Also, additional files, which may be needed to reproduce the results presented in the manuscript, are made available as supplementary material.
ZZ, XG and YW designed the project, YW and MJB collected the data, YW and XG performed the experiments and analyzed the data, YW, XG, MJB and ZZ drafted the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin. 2015;65(1):5–29.PubMedView ArticleGoogle Scholar
- Randall JM, Millard F, Kurzrock R. Molecular aberrations, targeted therapy, and renal cell carcinoma: current state-of-the-art. Cancer Metastasis Rev. 2014;33(4):1109–24.PubMedView ArticleGoogle Scholar
- Sato Y, Yoshizato T, Shiraishi Y, Maekawa S, Okuno Y, Kamura T, Shimamura T, Sato-Otsubo A, Nagae G, Suzuki H, et al. Integrated molecular analysis of clear-cell renal cell carcinoma. Nat Genet. 2013;45(8):860–7.PubMedView ArticleGoogle Scholar
- Guo X, Xu Y, Zhao Z. In-depth genomic data analyses revealed complex transcriptional and epigenetic dysregulations of BRAFV600E in melanoma. Mol Cancer. 2015;14:60.PubMedPubMed CentralView ArticleGoogle Scholar
- Jiang W, Jia P, Hutchinson KE, Johnson DB, Sosman JA, Zhao Z. Clinically relevant genes and regulatory pathways associated with NRASQ61 mutations in melanoma through an integrative genomics approach. Oncotarget. 2015;6(4):2496–508.PubMedView ArticleGoogle Scholar
- Eser S, Schnieke A, Schneider G, Saur D. Oncogenic KRAS signalling in pancreatic cancer. Br J Cancer. 2014;111(5):817–22.PubMedPubMed CentralView ArticleGoogle Scholar
- Croce CM. Oncogenes and cancer. N Engl J Med. 2008;358(5):502–11.PubMedView ArticleGoogle Scholar
- Gossage L, Eisen T, Maher ER. VHL, the story of a tumour suppressor gene. Nat Rev Cancer. 2015;15(1):55–64.PubMedView ArticleGoogle Scholar
- Brugarolas J. Molecular genetics of clear-cell renal cell carcinoma. J Clin Oncol Off J Am Soc Clin Oncol. 2014;32(18):1968–76.View ArticleGoogle Scholar
- Duns G, Hofstra RM, Sietzema JG, Hollema H, van Duivenbode I, Kuik A, Giezen C, Jan O, Bergsma JJ, Bijnen H, et al. Targeted exome sequencing in clear cell renal cell carcinoma tumors suggests aberrant chromatin regulation as a crucial step in ccRCC development. Hum Mutat. 2012;33(7):1059–62.PubMedView ArticleGoogle Scholar
- da Costa WH, Rezende M, Carneiro FC, Rocha RM, da Cunha IW, Carraro DM, Guimaraes GC, de Cassio ZS. Polybromo-1 (PBRM1), a SWI/SNF complex subunit is a prognostic marker in clear cell renal cell carcinoma. BJU Int. 2014;113(5b):E157–63.PubMedView ArticleGoogle Scholar
- Wilson BG, Roberts CW. SWI/SNF nucleosome remodellers and cancer. Nat Rev Cancer. 2011;11(7):481–92.PubMedView ArticleGoogle Scholar
- Cancer Genome Atlas Research N. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature. 2013;499(7456):43–9.View ArticleGoogle Scholar
- Xia W, Nagase S, Montia AG, Kalachikov SM, Keniry M, Su T, Memeo L, Hibshoosh H, Parsons R. BAF180 is a critical regulator of p21 induction and a tumor suppressor mutated in breast cancer. Cancer Res. 2008;68(6):1667–74.PubMedPubMed CentralView ArticleGoogle Scholar
- Huang L, Peng Y, Zhong G, Xie W, Dong W, Wang B, Chen X, Gu P, He W, Wu S, et al. PBRM1 suppresses bladder cancer by cyclin B1 induced cell cycle arrest. Oncotarget. 2015;6(18):16366–78.PubMedPubMed CentralView ArticleGoogle Scholar
- Ryan RJ, Bernstein BE. Molecular biology. Genetic events that shape the cancer epigenome. Science. 2012;336(6088):1513–4.PubMedView ArticleGoogle Scholar
- Wang Z, Zhai W, Richardson JA, Olson EN, Meneses JJ, Firpo MT, Kang C, Skarnes WC, Tjian R. Polybromo protein BAF180 functions in mammalian cardiac chamber maturation. Genes Dev. 2004;18(24):3106–16.PubMedPubMed CentralView ArticleGoogle Scholar
- Jiao Y, Pawlik TM, Anders RA, Selaru FM, Streppel MM, Lucas DJ, Niknafs N, Guthrie VB, Maitra A, Argani P, et al. Exome sequencing identifies frequent inactivating mutations in BAP1, ARID1A and PBRM1 in intrahepatic cholangiocarcinomas. Nat Genet. 2013;45(12):1470–3.PubMedPubMed CentralView ArticleGoogle Scholar
- Varela I, Tarpey P, Raine K, Huang D, Ong CK, Stephens P, Davies H, Jones D, Lin ML, Teague J, et al. Exome sequencing identifies frequent mutation of the SWI/SNF complex gene PBRM1 in renal carcinoma. Nature. 2011;469(7331):539–42.PubMedPubMed CentralView ArticleGoogle Scholar
- Benusiglio PR, Couve S, Gilbert-Dussardier B, Deveaux S, Le Jeune H, Da Costa M, Fromont G, Memeteau F, Yacoub M, Coupier I, et al. A germline mutation in PBRM1 predisposes to renal cell carcinoma. J Med Genet. 2015;52(6):426–30.PubMedView ArticleGoogle Scholar
- Kapur P, Pena-Llopis S, Christie A, Zhrebker L, Pavia-Jimenez A, Rathmell WK, Xie XJ, Brugarolas J. Effects on survival of BAP1 and PBRM1 mutations in sporadic clear-cell renal-cell carcinoma: a retrospective analysis with independent validation. Lancet Oncol. 2013;14(2):159–67.PubMedPubMed CentralView ArticleGoogle Scholar
- Gerlinger M, Horswell S, Larkin J, Rowan AJ, Salm MP, Varela I, Fisher R, McGranahan N, Matthews N, Santos CR, et al. Genomic architecture and evolution of clear cell renal cell carcinomas defined by multiregion sequencing. Nat Genet. 2014;46(3):225–33.PubMedPubMed CentralView ArticleGoogle Scholar
- Zhao M, Sun J, Zhao Z. TSGene: a web resource for tumor suppressor genes. Nucleic Acids Res. 2013;41(Database issue):D970–6.PubMedView ArticleGoogle Scholar
- Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26(1):139–40.PubMedView ArticleGoogle Scholar
- Safran M, Dalah I, Alexander J, Rosen N, Iny Stein T, Shmoish M, Nativ N, Bahir I, Doniger T, Krug H, et al. GeneCards Version 3: the human gene integrator. Database J Biol Databases Acuration. 2010;2010:baq020. doi:10.1093/database/baq020.Google Scholar
- Cooper SJ, Zou H, Legrand SN, Marlow LA, von Roemeling CA, Radisky DC, Wu KJ, Hempel N, Margulis V, Tun HW, et al. Loss of type III transforming growth factor-beta receptor expression is due to methylation silencing of the transcription factor GATA3 in renal cell carcinoma. Oncogene. 2010;29(20):2905–15.PubMedPubMed CentralView ArticleGoogle Scholar
- Tun HW, Marlow LA, von Roemeling CA, Cooper SJ, Kreinest P, Wu K, Luxon BA, Sinha M, Anastasiadis PZ, Copland JA. Pathway signature and cellular differentiation in clear cell renal cell carcinoma. PLoS One. 2010;5(5):e10696.PubMedPubMed CentralView ArticleGoogle Scholar
- Esteller M. CpG island hypermethylation and tumor suppressor genes: a booming present, a brighter future. Oncogene. 2002;21(35):5427–40.PubMedView ArticleGoogle Scholar
- Sethupathy P, Corda B, Hatzigeorgiou AG. TarBase: A comprehensive database of experimentally supported animal microRNA targets. RNA. 2006;12(2):192–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Hsu SD, Lin FM, Wu WY, Liang C, Huang WC, Chan WL, Tsai WT, Chen GZ, Lee CJ, Chiu CM, et al. miRTarBase: a database curates experimentally validated microRNA-target interactions. Nucleic Acids Res. 2011;39(Database issue):D163–9.PubMedView ArticleGoogle Scholar
- Xu Y, Guo X, Sun J, Zhao Z. Snowball: resampling combined with distance-based regression to discover transcriptional consequences of a driver mutation. Bioinformatics. 2015;31(1):84–93.PubMedView ArticleGoogle Scholar
- Xia J, Jia P, Hutchinson KE, Dahlman KB, Johnson D, Sosman J, Pao W, Zhao Z. A meta-analysis of somatic mutations from next generation sequencing of 241 melanomas: a road map for the study of genes with potential clinical relevance. Mol Cancer Ther. 2014;13(7):1918–28.PubMedPubMed CentralView ArticleGoogle Scholar
- Nam SJ, Lee C, Park JH, Moon KC. Decreased PBRM1 expression predicts unfavorable prognosis in patients with clear cell renal cell carcinoma. Urol Oncol. 2015;33(8):340 e349–16.View ArticleGoogle Scholar
- Hakimi AA, Ostrovnaya I, Reva B, Schultz N, Chen YB, Gonen M, Liu H, Takeda S, Voss MH, Tickoo SK, et al. Adverse outcomes in clear cell renal cell carcinoma with mutations of 3p21 epigenetic regulators BAP1 and SETD2: a report by MSKCC and the KIRC TCGA research network. Clin Cancer Res Off J Am Assoc Cancer Res. 2013;19(12):3259–67.View ArticleGoogle Scholar
- Ho TH, Kapur P, Joseph RW, Serie DJ, Eckel Passow JE, Parasramka M, Cheville JC, Wu KJ, Frenkel E, Rakheja D, et al. Loss of PBRM1 and BAP1 expression is less common in non-clear cell renal cell carcinoma than in clear cell renal cell carcinoma. Urol Oncol. 2015;33(1):23 e29–14.View ArticleGoogle Scholar
- Du P, Zhang X, Huang CC, Jafari N, Kibbe WA, Hou L, Lin SM. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics. 2010;11:587.PubMedPubMed CentralView ArticleGoogle Scholar
- Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, Ellis B, Gautier L, Ge Y, Gentry J, et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 2004;5(10):R80.PubMedPubMed CentralView ArticleGoogle Scholar
- Bindea G, Mlecnik B, Hackl H, Charoentong P, Tosolini M, Kirilovsky A, Fridman WH, Pages F, Trajanoski Z, Galon J. ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics. 2009;25(8):1091–3.PubMedPubMed CentralView ArticleGoogle Scholar
- Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–504.PubMedPubMed CentralView ArticleGoogle Scholar
- Benjamini Y, Hochberg Y. Controlling the False Discovery Rate - a Practical and Powerful Approach to Multiple Testing. J Roy Stat Soc B Met. 1995;57(1):289–300.Google Scholar
- Matys V, Fricke E, Geffers R, Gossling E, Haubrock M, Hehl R, Hornischer K, Karas D, Kel AE, Kel-Margoulis OV, et al. TRANSFAC: transcriptional regulation, from patterns to profiles. Nucleic Acids Res. 2003;31(1):374–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Cowley MJ, Pinese M, Kassahn KS, Waddell N, Pearson JV, Grimmond SM, Biankin AV, Hautaniemi S, Wu J. PINA v2.0: mining interactome modules. Nucleic Acids Res. 2012;40(Database issue):D862–5.PubMedView ArticleGoogle Scholar