The Ronglai Shen Lab

My research has focused on the development of novel statistical and bioinformatic methodologies that facilitate the study of translational research questions in cancer genomics and immuno-oncology. I developed several algorithms including iCluster and FACETS that are widely used in research community. My current research program focuses on two branches. First, in collaboration with Colin Begg, we are investigating the information content from rare somatic variants in the cancer genome toward identifying the primary for cancers detected via ctDNA or for cancers of unknown primary. Our group developed methods that made use of concepts developed in ecology and natural language processing to mine these rare signals across the whole genome for tracing the cancer site of origin. We are further extending these methods to identify somatic-germline associations in population-scale genome sequencing studies. In the second research branch, with a close collaboration with Margaret Callahan, a melanoma medical oncologist, we are mining multimodal datasets including flow cytometry, CITE-seq, multiplexed immunofluorescent imaging, and spatial transcriptomics to study diverse immune cell types in patient blood and tissue samples that predicts treatment response to immunotherapy. Our recent work led to the discovery of a blood-based LAG+ immunotype as an underlying mechanism of treatment resistance to anti-PD-1 therapy. Using a topic model, we linked this immunotype with an exhaustion phenotype that lacks immune-activation upon treatment using patient samples collected from a recent clinical trial. We are employing multi-modal single cell sequencing techniques to further characterize the LAG+ cells and spatial imaging technologies to study the characteristics of the tumor microenvironment.