Guoshuai Cai ColloquiumTuesday, March 1, 2016 - 2:45pm
Statistics Department Colloquium
Where: LeConte College, Room 210
Speaker: Guoshuai Cai
Affiliation: Dartmouth College, Department of Genetics
Title: Differential Expression and Methylation-expression Integration Analyses in RNA-seq
Abstract: High throughput and high dimensional data are increasingly generated in cancer study and cause two challenges in the analysis, identification of relevance from global analyses and integration of various types of biological data. In this talk, two methods will be discussed as responses to these two challenges in analyzing data from RNA-seq, which is becoming a common technique for surveying gene expression. As it is still unclear how the increased sequencing capacity has affected measurement accuracy of mRNA, we sought to investigate their relationship in the first project. We found that measurement accuracy improves with sequencing depth. Based on our discovery, we proposed a novel model that can better capture the overdispersion in the real data. The second method we developed was to identify methylation regulated genes by integrating RNA-seq and DNA methylation datasets. The specific effective thresholds for methylation were determined and then utilized to dichotomize the methylation status, which were further applied in tumor subtyping. With application to breast cancer, we identified a new methylator phenotype related with malignant clinical features. This study provides the valuable basis for differential expression detection and cancer epigenetics using high-throughput transcriptome profiling data.