Insights into RNA Biology from Statistical Modeling for CancerTherapeutics
Cancer is often associated with aberrant gene expression at the post-transcriptional level. However, it has not been fully understood how post-transcriptional regulation alters gene expression for cancer. Since hundreds RNAs interact with other hundreds RNAs simultaneously at the level, their tumorigenic mechanisms need to be understood in consideration of the interactions.
In my talk, I will present statistical/computational models successfully identifying tumorigenic mechanisms and functions in the post-transcriptional level. First, I will identify a tumorigenic mechanism of 3'UTR shortening - a type of post-transcriptional regulation - widespread in human cancers by considering their interactions with other types of RNAs in integrated network modeling. Second, I will introduce a signal-enrichment method identifying tumorigenic functions of the interactions in terms of biological pathways. Further, I will show statistical methods and data mining techniques identifying the translational implications of the interactions using publicly available big data.
As interactions among RNA species become an important issue in delineating RNA biology, computational models shed new insights into its functions in human diseases.