Our long term goal is to integrate structure and sequence based approaches founded in statistical mechanics to understand key features of molecular recognition by proteins, as well as protein fitness and function more generally. 1. Mapping Complex Conformational and Fitness Landscapes of Proteins Conformational dynamics plays a fundamental role in the regulation of molecular recognition and statistical mechanics provides the framework to derive a comprehensive theory for the binding free energy of a ligand to a protein. Our goal is to use advanced sampling methods based on molecular dynamics simulations to construct conformational free energy landscapes of sufficient accuracy to be predictive for thermodynamic and kinetic properties, but also as important, to generate qualitative insights about the molecular mechanisms for binding and allosteric conformational transitions. Powerful inverse inference statistical approaches are being developed to study the relationship between protein sequence co-variation and protein fitness. The co- variation of pairs of mutations contained in multiple sequence alignments of protein families will be used to build Potts Hamiltonian models of the sequence patterns that can be used to predict the change in fitness resulting from drug selection pressure, as well as infer features of the conformational propensities of individual proteins. 2. The Structural Basis for Kinase Selectivity and Regulation by Small Molecules The human kinome encodes about 518 kinases (PKs) which constitute one of the largest class of genes. Progress in kinase structural biology offers a conceptual framework for understanding many aspects of kinase biology. With our collaborators at the Fox Chase Cancer Center and Columbia University we are working on biophysical simulation and evolutionary sequence based approaches to rationalize biochemical profiling studies of kinases and to devise a framework for understanding the molecular mechanisms of selectivity of kinase inhibitors to their targets. 3. Inhibition of HIV-1 Proteins and Mechanisms of Drug Resistance In collaboration with groups at the University of Colorado, Harvard and Scripps, I am working on the allosteric basis for inhibition by small molecules of HIV-1 proteins, on mechanisms of drug resistance, and on comparative studies of the fitness of HIV proteins in different HIV clades. Allosteric HIV-1 IN inhibitors called ALLINIs are an important new class of anti-HIV-1 agents. ALLINIs bind at the IN catalytic core domain (CCD) dimer interface occupying the principal binding pocket of LEDGF. Using our conformational free energy simulation tools and the sequence based tools we are developing to understand correlated mutations, we are working with our collaborators to ascertain the inhibitory mechansims of ALLINIs, and the basis for drug resistance.
The conformational free energy and fitness landscapes of proteins reveal fundamental features about the relationship between protein sequence, structure, and function. Our goal is to develop models of the conformational landscapes of proteins to be predictive for their thermodynamic, and kinetic properties, and to develop models of the fitness landscapes of proteins from sequence co-variation data that can provide insights into how proteins respond to drug selection and other environmental pressures. With our collaborators, we are working on structure based inhibitor design and the acquisition of resistance mutations in HIV-1 proteins, and on biophysical simulation and evolutionary sequence based approaches to rationalize biochemical profiling studies of kinase family proteins.
