Multivariate Bioinformatics Analysis of Taiwanese Cancer RNA-Seq Data to Study Human Endogenous Retroviruses as Carcinogens

The more we understand the cancer causes and risk factors, we are better able to provide cancer treatment therapies that specifically target the causing molecules. The common risk factors include family history, older age, smoking, poor diet and pathogen infections. Although we put a lot of efforts and budgets on cancer research, cancer is just as deadly as 50 years ago. Therefore it is required that we have to explore with different scope and models in cancer studies. Along with the rapid development of biotechnologies, such as next generation sequencing (NGS), scientists realized that the once thought “junk DNA” is actively transcribed up to 93% of human genome, and now called non-coding RNA. Among them, Human Endogenous Retrovirus (HERV) sequences, accounting for 8% of our entire genome and having invaded our genome at least 25 million years ago, might be one of the new risk factors in cancer. We have shown with evidence that HERVs indeed associated with cancers based on our previous studies using public RNA-seq data. Here in this two-year proposal we propose to use a total of 36 RNA-seq big data derived from 3 different Taiwanese cancers, each with 6 matched cancerous and normal control biopsies to screen the HERV genome. We will use a diversity of bioinformatics tools to address the insertional polymorphism of HERVs between Taiwanese and Caucasian diseases, the functions of differentially expressed HERV peptides in cancers, the protein-protein interactions of HERV and host proteins, interactions between endogenous and exogenous virus proteins. The annotation and interpretation results will improve our understanding in cancers and might significantly change our targeted cancer therapies in the future.