TTSurv: Exploring the Multi-Gene Prognosis in Thousands of Tumors

Abstract

Thoracic malignancies are a common type of cancer and area major global health problem. These complex diseases, including lung cancer, esophageal cancer, and breast cancer, etc. have attracted considerable attention from researchers. Potential gene-cancer associations can be explored by demonstrating the association between clinical data and gene expression data. Emerging evidence suggests that the transcriptome plays a particularly critical role as a diagnostic biomarker in pathology and histology studies. Thus, there is an urgent need to develop a platform that allows users to perform a comprehensive prognostic analysis of thoracic cancers. Here, we developed TTSurv, which aims to correlate coding and noncoding genes with cancers by combining high-throughput data with clinical prognosis. TTSurv focuses on the application of high-throughput data to detect ncRNAs, such as lncRNAs and microRNAs, as novel diagnostic and prognostic biomarkers. For a more comprehensive analysis, a large amount of public expression profile data with clinical follow-up information have been integrated into TTSurv. TTSurv also provides flexible methods such as a minimum p-value algorithm and unsupervised clustering methods that can classify thoracic cancer samples into different risk groups. TTSurv will expand our understanding of ncRNAs in thoracic malignancies and provide new insights into their application as potential prognostic/diagnostic biomarkers.