The landscape of antisense gene expression in human cancers

Genome Res. 2015 Jul;25(7):1068-79. doi: 10.1101/gr.180596.114. Epub 2015 Jun 10.

Abstract

High-throughput RNA sequencing has revealed more pervasive transcription of the human genome than previously anticipated. However, the extent of natural antisense transcripts' (NATs) expression, their regulation of cognate sense genes, and the role of NATs in cancer remain poorly understood. Here, we use strand-specific paired-end RNA sequencing (ssRNA-seq) data from 376 cancer samples covering nine tissue types to comprehensively characterize the landscape of antisense expression. We found consistent antisense expression in at least 38% of annotated transcripts, which in general is positively correlated with sense gene expression. Investigation of sense/antisense pair expressions across tissue types revealed lineage-specific, ubiquitous and cancer-specific antisense loci transcription. Comparisons between tumor and normal samples identified both concordant (same direction) and discordant (opposite direction) sense/antisense expression patterns. Finally, we provide OncoNAT, a catalog of cancer-related genes with significant antisense transcription, which will enable future investigations of sense/antisense regulation in cancer. Using OncoNAT we identified several functional NATs, including NKX2-1-AS1 that regulates the NKX2-1 oncogene and cell proliferation in lung cancer cells. Overall, this study provides a comprehensive account of NATs and supports a role for NATs' regulation of tumor suppressors and oncogenes in cancer biology.

MeSH terms

  • Cluster Analysis
  • Computational Biology / methods
  • Gene Expression Profiling
  • Gene Expression Regulation, Neoplastic*
  • Genetic Loci
  • Humans
  • Neoplasms / genetics*
  • Organ Specificity / genetics
  • RNA, Antisense / genetics*
  • Transcriptome*

Substances

  • RNA, Antisense

Associated data

  • GEO/GSE66729
  • SRA/SRP048484