Physical simulations, such as molecular dynamics of atomistic forcefields, are capable, in principle, of determining free energies of different conformations. But, it is computationally expensive for many interesting problems that entail large proteins or large actions. We recently developed the Modeling Employing Limited Data (MELD) method, that allows is to combine sparse, ambiguous, or uncertain information into simulations without breaking detailed balance. This greatly reduces the phase space that needs to be sampled, speeding up the computations of conformational free energies. For example, we used MELD to compete in the protein-structure prediction competition called CASP, and to predict relative binding free energies of peptides and small drug-like molecules to a proteins.

Using information to reduce the phase space of molecular dynamic simulations of biologically relevant problems

255th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.