Computational Approaches for Predicting Transcription Factor Interactions

Computational Approaches for Predicting Transcription Factor Interactions

Computational Approaches for Predicting Transcription Factor Interactions

Wednesday, November 9, 2011 - 4:00pm

The Wistar Institute, 3601 Spruce Street, Philadelphia, PA

Computational Approaches for Predicting Transcription Factor Interactions

Mona Singh, Ph.D.
Princeton University

From Mona Singh:

Proteins accomplish virtually all of their cellular functions via interactions with other molecules.  As a result, a broad array of computational methods have been developed to predict protein interactions, whether with DNA, other proteins, or small molecules. First, I will describe a general framework we have developed for de novo prediction of protein-DNA (and protein-protein) interactions mediated by specific structural interfaces, and demonstrate its applicability on Cys2His2 zinc fingers, the largest class of transcription factors in eukaryotic genomes. Next, I will describe a genomic approach for uncovering high-confidence regulatory interactions for transcription factors with known specificity. Finally, I will describe highly effective sequence- and structure-based approaches my group has developed for predicting sites in protein sequences that interact with small molecules.