1612 MICRORNA EXPRESSION ANALYSIS REVEALS SIGNIFICANT BIOLOGICAL PATHWAYS IN HUMAN PROSTATE TUMOR

Bairong Shen,Yifei Tang

DOI: https://doi.org/10.1016/j.juro.2011.02.1720

2011-01-01

Abstract:You have accessJournal of UrologyProstate Cancer: Basic Research1 Apr 20111612 MICRORNA EXPRESSION ANALYSIS REVEALS SIGNIFICANT BIOLOGICAL PATHWAYS IN HUMAN PROSTATE TUMOR Bairong Shen and Yifei Tang Bairong ShenBairong Shen Suzhou, China, People's Republic of More articles by this author and Yifei TangYifei Tang Suzhou, China, People's Republic of More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2011.02.1720AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES MicroRNAs (miRNAs) are reported to play important roles in tumor formation. Microarray technologies are applied to study the microRNA expression profile in cancer. However, it is very common that the set of differentially expressed miRNAs (DE-miRNA) related to the same tumor identified from different laboratories varies widely. Our present work aims to identify common prostate cancer associated pathways from the varied differentially expressed miRNAs sets. METHODS We collected 4 independent miRNA profiling datasets from Gene Expression Omnibus (GEO), which are of four platforms with miRNA probes designed using Sanger miRBase release. We preprocessed those datasets and performed differential expression analysis by Cancer Outlier Profile Analysis (COPA), a simple method based on robust centering and scaling of the data. The target genes of these DE-miRNAs were retrieved from both publicly available TargetScanHuman database and the overlapped result of three commonly used bioinformatics tools (miRanda, RNAhybrid, and TargetSpy). DE-miRNAs were then mapped to GO, GeneGO and KEGG databases for enrichment analysis. Overlapping analysis was performed at miRNA and pathway levels to test our hypothesis that consistency of miRNA profiling experiments at pathway level is expected to be higher than that at miRNA level. RESULTS In this study, we processed four miRNA profiling datasets generated from human prostate tissues by the meta-analysis method. 155 DE-miRNAs and 9114 putative target genes, 742 Molecular Function terms, 3266 Biological Process terms and 424 GeneGO Pathways were identified with P-value < 0.05. By the overlapping analysis, all the six pair-wise comparison showed that the consistency of DE-miRNA profiling at pathway level is significantly higher than that at miRNA level. CONCLUSIONS We identified 155 DE-miRNAs related to prostate tumor, some of them are experimentally reported and the others related to the enriched pathways common to all datasets deserve further experimental verification. In addition, we also confirmed that the consistency of prostate tumor expression profiles at pathway level is higher than that at microRNA level. © 2011 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 185Issue 4SApril 2011Page: e646-e647 Advertisement Copyright & Permissions© 2011 by American Urological Association Education and Research, Inc.Metrics Author Information Bairong Shen Suzhou, China, People's Republic of More articles by this author Yifei Tang Suzhou, China, People's Republic of More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...