My primary research is driven by the need for powerful statistical methods to address the challenges those technologies have posed for data analysis and interpretation, particularly for data emerging from biological and biomedical studies, such as epigenetic and cancer genomics related research. I have developed novel methodologies for a variety of problems, including change point detection methods for identifying somatic copy number aberration, nonparametric Bayesian methods to integrate the heterogeneity in somatic mutations into gene expression analysis, Gaussian graphical models for eQTL analysis and methods for the analysis of single cell sequencing data. My ultimate goal is to develop methods that can integrate genomic features into the prediction of clinical outcomes, which will potentially shed new lights on personalized disease diagnosis and prognosis.