Rational protein design.

We aim for the computational design of biomolecular interactions. Two types of biomolecular interactions will be studied: protein-protein interactions and protein-ligand interactions. In the case of protein-protein interactions, ubiquitin, a protein with promiscuous binding activity, will be used as target. This study will serve as a proof of concept that by changing the dynamics, rather than the interacting surface, the specificity of protein-protein interactions can be tuned. In the case of protein-ligand interactions the aim is to design proteins that bind and metabolize novel substrates. Here, nucleic acid polymerases accepting substrates with different leaving groups (i.e. the pyrophosphate moiety is replaced by another group) will be designed. At the core of the design protocol are physics based free energy calculations, taking the full system and its natural environment into account. Preliminary calculations proved to be very accurate, and are now feasible at the required scale with efficient software, automated algorithms and state-of-the-art hardware. To cope with the combinatorial explosion in case of multiple mutations, a protocol is proposed making use of quick filters and state-of-the-art bio-informatics tools. This results in a small (< 100 mutants) highly enriched library that will be tested experimentally.

Keywords:Computational biophysics, Protein design, Molecular simulations, Bioinformatics

Disciplines:Inorganic chemistry, Organic chemistry, Theoretical and computational chemistry, Other chemical sciences