BioMaPS Institute for Quantitative Biology
Rutgers - The State University of New Jersey
Large-scale prediction, characterization and modulation of protease enzyme specificity using computation and experiment
Characterizing the substrate specificity of protease enzymes is critical for illuminating the molecular basis of their diverse and complex roles in a wide array of biological processes. Rapid and accurate prediction of their extended substrate specificity would also aid in the design of custom proteases capable of selectively and controllably cleaving biotechnologically or therapeutically relevant targets. A catalytic drug such as a programmed protease would have several advantages over binding-based moieties such as antibodies. Proteases are multispecific enzymes that cleave multiple substrates of disparate sequence while not cleaving other sequences, thereby showing signs of both positive and negative selection. Thus, the specificity landscape of proteases determines their functions, and will be key to the use of designed therapeutic proteases to ensure proper targeting and minimizing side-effects. We have developed a specificity modeling and design framework by combining in silico structure-based modeling using the Rosetta macromolecular modeling approach and machine learning with experimental in vivo assays and Deep Sequencing to elucidate and predictively modulate specificity landscapes of proteases on a large scale. In our framework tens of thousands of substrates are experimentally evaluated and this information is used to guide computational design approaches to make predictions for the entire cleavage specificity landscape (millions of substrate sequences). I will describe the framework and its application for uncovering and designing the specificity landscape of the Hepatitis C virus protease, and its drug-resistant mutants.