Eugene I. Shakhnovich
Department of Chemistry & Chemical Biology
Mallinckrodt Building, Room 106
12 Oxford Street, Cambridge, MA 02138
tel: (617) 495-4130; fax: (617) 384-9228
The protein-folding problem of how proteins find their native functionally active conformation is one of the most important and most difficult problems in modern biophysics. The difficulty of the problem can come from two sources (i) finding of free energy minimum is a very difficult optimization problem and (ii) energy potential may be not sufficiently accurate to distinguish between the native state and plethora of non-native conformations. We were able to overcome the last difficulty by development the algorithm, which, for any given conformation, allows us to design a sequence which has deep global minimum of energy in that conformation. The most important fact is that sequences designed in such a way indeed fold fast into their native conformations even if they are up to 100 monomers long! This approach allowed to overcome the major difficulties and provided with a model where folding can be studied rigorously and completely. Such features as nucleation-growth mechanism, character of intermediates and kinetic traps were studied in detail recently. The obtained advanced understanding of protein folding mechanisms allowed to make an intelligent connection with molecular evolution and helped to discern the conservatism patterns in protein families that are responsible for function, protein stability and folding kinetics. Better understanding of molecular evolution, allowed, in turn to make detailed predictions for protein engineering experiments. Such experiments carried out by us on villin and by others on several other proteins provided a major verification of the evolving understanding of the protein folding mechanism(s).
Protein design direction is being pursued in the group with aspects related to design of stable and fast folding proteins as well as design of peptide and non-peptide ligands to pharmaceutically relevant proteins with known structure.
Another important direction of our group is a physical approach to bioinformatics, i.e. making predictions of protein fold and function from sequence, based on physical mechanisms of protein folding and fold recognition. That also allows us to address such important topics in bioinformatics as identification of ORFs and introns.
Mirny LA, Shakhnovich EI. (1998). Protein structure prediction by threading. Why it works and why it does not. Jnl Molec. Biol. 283: (2) 507-526.
DeWitte RS, Ishchenko AV, Shakhnovich EI. (1997). SMoG: De novo design method based on simple, fast, and accurate free energy estimates .2. Case studies in molecular design. J Am Chem Soc 119: (20) 4608-4617.
Shakhnovich EI, Abkevich V, Ptitsyn O. (1996). Conserved residues and the mechanism of protein folding. Nature 379: (6560) 96-98.
Page created and maintained by Xaq Pitkow