Computational Chemical and Structural Biology
Marcin J. Skwark
Current Post-Doctoral Fellow
Ph.D Theoretical Chemistry, Stockholm University
firstname.middleinitial.lastname [at] vanderbilt.edu
Protein structure prediction ab-initio is one of the longest standing challenges in structural biology. An universally applicable, accurate solution to the problem promises advantages spanning several research fields, such as drug discovery, interpretation of functional studies, experiment design and protein design. Regretfully, the search space for conformational exploration grows exponentially with the length of the protein, rendering ab-initio methods unfeasible for all, but relatively short protein chains. One approach to enhance the structure prediction of a protein is to predict interacting residues for use as constraints during the folding process, thus effectively limiting the size of conformational search space. Such an approach is not only theoretically possible, but due to the recent developments in the field of contact prediction, feasible for the rapidly growing number of protein families. In my research, I focus on developing methods, that unify statistical coupling inference from large multiple sequence alignments with effective conformational sampling, allowing for successful prediction of previously unknown protein structures.