Figure1 revised

GSE66354

GSE68015

Limma

RobustRankAggreg

Pathological analysis

A total of 22 candidate prognostic genes were identified.

Clinical data

2369DEGs

GSE16011

Extract duplicates

RNA sequencing data

2671DEGs

GSE94349

GEPIA (http://gepia.cancer-pku.cn/index.html)

GSE61335

KEGG pathway enrichment analysis

GSE86574

ldentifying2739DEGs(962 down-regulated and1777up-regulated genes) from TCGA

4068DEGs

Prognostic  analysis

Nine datasets from GEO(A total of 601 samples)

LASSO regression

A total of 512 overlapping DEGs (212 down regulated and 300 up regulated genes )were identified by intersecting the results of integrated microarray and RNA sequencing data analysis

TCGA clinical data(528 samples)

STRING&Cytoscape PPI Network construction and Module Analysis

GSE111260

Gene Ontology(GO) Term enrichment analysis

GSE50161

1938DEGs

1816DEGs

2727DEGs

ldentifying 670 DEGs( 293 down-regulated and 377 up regulated genes) from 9 GEO datasets

GSE15824

Venn Diagram

3325DEGs

Univariate Coxproportional hazards regression analysis

Multivariate  Coxproportional hazards regression analysis

1416DEGs

2990DEGs