Weighted gene co-expression network analysis of gene modules for the prognosis of esophageal cancer

Abstract

Esophageal cancer is a common malignant tumor, whose pathogenesis and prognosis factors are not fully understood. This study aimed to discover the gene clusters that have similar functions and can be used to predict the prognosis of esophageal cancer. The matched microarray and RNA sequencing data of 185 patients with esophageal cancer were downloaded from The Cancer Genome Atlas (TCGA), and gene co-expression networks were built without distinguishing between squamous carcinoma and adenocarcinoma. The result showed that 12 modules were associated with one or more survival data such as recurrence status, recurrence time, vital status or vital time. Furthermore, survival analysis showed that 5 out of the 12 modules were related to progression-free survival (PFS) or overall survival (OS). As the most important module, the midnight blue module with 82 genes was related to PFS, apart from the patient age, tumor grade, primary treatment success, and duration of smoking and tumor histological type. Gene ontology enrichment analysis revealed that “glycoprotein binding” was the top enriched function of midnight blue module genes. Additionally, the blue module was the exclusive gene clusters related to OS. Platelet activating factor receptor (PTAFR) and feline Gardner-Rasheed (FGR) were the top hub genes in both modeling datasets and the STRING protein interaction database. In conclusion, our study provides novel insights into the prognosis-associated genes and screens out candidate biomarkers for esophageal cancer.

Keywords

esophageal cancer / The Cancer Genome Atlas / co-expression network analysis / weighted gene co-expression network analysis / enrichment analysis

References

[1] ChenW, ZhengR, ZengH, et al. . Annual report on status of cancer in China, 2011. Chin J Can Res (Chinese), 2015, 27(1): 2-12

[2] BlotWJ, McLaughlinJK. The changing epidemiology of esophageal cancer. Semin Oncol, 1999, 26: 2-8 PMID: 10566604

[3] LangfelderP, HorvathS. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics, 2008, 9: 559 PMID: 19114008 PMCID: 2631488

[4] ZhangJ, BaddooM, HanC, et al. . Gene network analysis reveals a novel 22-gene signature of carbon metabolism in hepatocellular carcinoma. Oncotarget, 2016, 31(7): 49232-49245

[5] PressonAP, YoonNK, BagryanovaL, et al. . Protein expression based multimarker analysis of breast cancer samples. BMC Cancer, 2011, 11: 230 PMID: 21651811 PMCID: 3142534

[6] UptonA, ArvanitisTN. Using evolutional properties of gene networks in understanding survival prognosis of glioblastoma. IEEE T Inf Technol B, 2014, 18(3): 810-816

[7] ZhaoX, CaiH, WangX, et al. . Discovery of signature genes in gastric cancer associated with prognosis. Neoplasma, 2016, 63(2): 239-245 PMID: 26774142

[8] JiaX, MiaoZ, LiW, et al. . Cancer-risk module identification and module-based disease risk evaluation: a case study on lung cancer. PloS one, 2014, 9(3): 92395

[9] WanYW, AllenGI, LiuZ. TCGA2STAT: simple TCGA data access for integrated statistical analysis in R. Bioinformatics, 2016, 32(6): 952-954 PMID: 26568634

[10] HuberW, CareyVJ, GentlemanR, et al. . Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods, 2015, 12(2): 115-121 PMID: 25633503 PMCID: 4509590

[11] Hastie T, Narasimhan B, Chu G. Impute: Imputation for microarray data. In R package version 1.46.0, 2016

[12] Kosinski M, Biecek P. RTCGA: The cancer genome atlas data integration. In R package version 1.2.2, 2015

[13] LangfelderP, HorvathS. Fast R functions for robust correlations and hierarchical clustering. J Stat Softw, 2012, 46(11): i11 PMID: 23050260 PMCID: 3465711

[14] LangfelderP, HorvathS. Eigengene networks for studying the relationships between co-expression modules. BMC Syst Biol, 2007, 1: 54 PMID: 18031580 PMCID: 2267703

[15] Lopez-RatonM, Rodriguez-AlvarezMX, Cadarso-SuarezC, et al. . Optimal cutpoints: An R package for selecting optimal cutpoints in diagnostic tests. J Stat Softw, 2014, 61(8): 1-36

[16] HuZ, HungJH, WangY, et al. . VisANT 3.5: multi-scale network visualization, analysis and inference based on the gene ontology. Nucleic Acids Res, 2009, 37: W115-W121 PMID: 19465394 PMCID: 2703932

[17] JensenLJ, KuhnM, StarkM, et al. . STRING 8-a global view on proteins and their functional interactions in 630 organisms. Nucleic Acids Res, 2009, 37: D412-D416 PMID: 18940858

[18] JacomyM, VenturiniT, HeymannS, et al. . ForceAtlas2, a continuous graph layout algorithm for handy network visualization designed for the Gephi software. PloS one, 2014, 9(6): 98679

[19] FangH, GoughJ. The ‘dnet’ approach promotes emerging research on cancer patient survival. Genome Med, 2014, 6(8): 64 PMID: 25246945 PMCID: 4160547

[20] ChasePB, YangJM, ThompsonFH, et al. . Regional mapping of the human platelet-activating factor receptor gene (PTAFR) to 1p35->p34.3 by fluorescence in situ hybridization. Cytogenet Cell Genet, 1996, 72(2–3): 205-207 PMID: 8978777

[21] AgrawalV, JaiswalMK, IlievskiV, et al. . Platelet-activating factor: a role in preterm delivery and an essential interaction with Toll-like receptor signaling in mice. Biol Reprod, 2014, 91(5): 119 PMID: 25253732 PMCID: 4434927

[22] WangQ, AnY, YuanQ, et al. . Identification of allelic expression imbalance genes in human hepatocellular carcinoma through massively parallel DNA and RNA sequencing. Med Oncol, 2016, 33(4): 38 PMID: 27000824

[23] KimHS, HanHD, Armaiz-PenaGN, et al. . Functional roles of Src and Fgr in ovarian carcinoma. Clin Can Res, 2011, 17(7): 1713-1721

[24] HuY, LiuY, PelletierS, et al. . Requirement of Src kinases Lyn, Hck and Fgr for BCR-ABL1-induced B-lymphoblastic leukemia but not chronic myeloid leukemia. Nat Genet, 2004, 36(5): 453-461 PMID: 15098032

[25] LiuY, BezverbnayaK, ZhaoT, et al. . Involvement of the HCK and FGR src-family kinases in FCRL4-mediated immune regulation. J Immunol, 2015, 194(12): 5851-5860 PMID: 25972488 PMCID: 4456631