Our lab uses computational approaches to study the genetics of human diseases, including cancer. A primary focus of our research is to develop novel tools for mapping risk genes of complex diseases from data of GWAS, family studies or somatic mutations in the case of cancer. We are also interested in related questions, such as how to predict functional significance of DNA mutations; how genes and environmental factors together influence the disease risks and what is the role of dysregulation of gene expression in diseases.

We develop and employ computational or statistical tools to address our questions, and work closely with geneticists and experimental biologists. A key feature of our strategy is the integration of multiple genomic datasets, such as transcriptome data, epigenetic data, and biological networks. This integrated approach could combine signals in different datasets to increase the power of studies. Furthermore, by putting DNA variations in the context of gene interaction and regulatory networks, it is possible to better understand the mechanism connecting genetic changes to phenotypes.

We are also interested in computational questions in regulatory genomics. How do cis-regulatory sequences interpret the information in cellular environments to drive spatial-temporal gene expression patterns? How do regulatory sequences change during evolution? We believe a better understanding of these questions will also help the study of human genetics, specifically by improving our ability to interpret variations in non-coding sequences.