Computational regulatory genomics
1. Computational cancer epigenomics:
I am interested in understanding how cancers commandeer the normal regulatory machinery to create disease. As a model system, I use Ewing sarcoma, a pediatric tumor, which is a good model because it is almost always driven by a single, well-characterized mutagenic event: a chromosomal translocation leading to the fusion protein EWS-FLI1. To explore how this fusion protein re-wires the cells to proliferate uncontrollably, I am examining genome-wide epigenetic profiles of Ewing sarcoma. These biological questions lead to computational problems inherent in dealing with lots of data from different individuals, cancers, and data types.
2. Single-cell sequencing analysis:
In the past, we have only been able to sequence populations of cells, leaving important cell-to-cell differences unexplored. New microfluidics and sequencing technology is making it possible to ask questions about single cells. I use this technology to ask fundamental questions about how cells differentiate and respond to their environments at the single cell level.
3. Gene regulation and chromatin structure:
I am interested in how cells fold their DNA to enable complex regulatory patterns. Humans are made up of many different cell-types. Though these cell-types share a single genome, they have very different phenotypes and functions. The basis for these dynamics is regulatory DNA, which governs when and where different genes are expressed. I use computational approaches like machine learning, supercomputing, and software engineering to analyze data from high-throughput ChIP-seq, DNase-seq, and ATAC-seq experiments to understand how cells change during development. I am also interested in the evolutionary background of regulatory differences.