Research Statement – Mark D .

As a computer scientist and interdisciplinary researcher working in the field of computational biology, I am inspired by the potential benefits of computational approaches to biological research and driven by my passion for solving difficult computational challenges. In that vein, the majority of my research has been in cancer genomics, which provides diverse and difficult computational problems with important applications. The cost of DNA sequencing is declining rapidly such that measuring a tumor’s mutations is sometimes cheaper than drug regimens, providing the opportunity for personalized cancer therapy tailored to a tumor’s mutation profile. However, a major challenge to developing new personalized therapies for cancer is the size and complexity of the datasets: The number of hypotheses far outstrips the resources to test them experimentally. Thus, computational approaches that automatically generate and prioritize hypotheses are an urgent priority. These approaches need to be computationally efficient – scaling to increasingly large cancer genomics datasets – and statistically rigorous in making predictions. The main problem I have focused on in my research is distinguishing the few causal mutations responsible for cancer from the numerous random mutations in a cohort of tumors. The ability to identify these key mutations has many benefits, including for identifying new drug targets and personalizing treatment. I have developed computational approaches for this problem – utilizing techniques from graph theory, statistics, and combinatorial optimization – and applied these approaches in multiple collaborations with cancer biologists. Looking forward, I am well positioned to continue to make an impact in cancer genomics. Already, advances in cancer research powered by DNA sequencing have led to new research questions and computational challenges. The expertise I have developed in cancer genomics has prepared me to continue and expand on my current research to face these challenges. Furthermore, new biotechnology and initiatives promise to generate datasets that can be used to address key open biological questions. As an interdisciplinary researcher with experience applying my methods to real datasets, I look forward to to collaborating with biologists to address computational challenges presented by these new cutting edge technologies.