Structure-based deep learning studies of transcription factor–DNA binding

Research in molecular biology has become increasingly quantitative and computational. This trend is due to two major drivers: Biology now generates large amounts of data in every experiment, and the power of computers has grown exponentially. The combination of data and computing is the basis for the recent development of AI-based computational biology methods. Over the years, my lab developed machine learning approaches to advance our understanding of gene regulation. Our studies focused on mechanistic insights of transcription factor-DNA binding specificity. This includes DNA shape recognition and protein-DNA binding. In my talk, I will discuss how a synthesis of computational methods and experimental data enabled insights in gene regulatory mechanism. I will further briefly introduce additional tools that my lab developed for the three-dimensional structure analysis of other biomolecules.