Machine learning in drug discovery
The focus of the research team is on developing novel machine learning algorithms for critical problems in drug discovery.
Main Research Directions
- Identification of genomically / transcriptomically defined cancer subtypes, and prediction of targeted therapeutics. Initially, the focus will be on acute myeloid leukemia and melanoma, on which we are collaborating with world-class experts.
- Selection of informative experiments is a key problem in biomedical sciences and we are investigating automated hypothesis generation, utilizing the significant theoretical work in active learning over the last decade.
- Drug-target interaction prediction is an important problem with applications to repurposing, mechanism identification and side effect prediction. We are working on novel algorithms that can address the challenges faced by current methods.
The ideal candidate will have strong command of statistics and machine learning. Fundamental knowledge of programming (Python/MATLAB) for translating ideas to working code is expected. Working knowledge of fundamental biological concepts would be beneficial, however, it is not a hard constraint.
University of Texas Southwestern Medical Center is a world-class powerhouse in biomedical sciences. The faculty includes six Nobel Laureates, four of whom are active faculty members, 22 members of the prestigious National Academy of Sciences. Investigations into cancer, neuroscience, heart disease and stroke, arthritis, diabetes, and many other fields keep UT Southwestern at the forefront of medical progress. The Harold C. Simmons Cancer Center of UT Southwestern has been designated by the National Cancer Institute (NCI) as a comprehensive center, an elite distinction held by only the top-tier cancer centers nationwide. The Lyda Hill Department of Bioinformatics has recently been established with an exceptional $25 million gift from Lyda Hill. The new Department is dedicated to developing UT Southwestern’s capability in bioinformatics.
Interested candidates should contact the principal investigator via e-mail with their CV.
Murat Can Cobanoglu, Ph.D.