COMPUTER SIMULATIONS OF PROTEIN FOLDING
Papoian, Garegin A
MetadataShow full item record
Understanding how proteins fold and interact with each other is key to understanding virtually all biological processes. Recent advances in computer power and modeling techniques make it possible to study proteins and other microscopic systems on biologically relevant time and length scales, closing the gap between simulations and experiments. At the same time, the emergence of more accurate models, derived from more rigorous physical principles, allows us to address a number of fundamental questions. The present work relies on molecular dynamics (MD) simulations to investigate several important aspects of protein behavior. First, we introduce the associative memory, water mediated, structure and energy model (AWSEM) and demonstrate its structure prediction capabilities. AWSEM is a coarse-grained protein force field that consists of many physically motivated potentials and a bioinformatically based term, which accounts for many-body local effects by matching its short sequential fragments to the sequences of experimentally resolved structures. We show that the AWSEM force field can be used for <italic>de novo</italic> structure prediction, as well as for kinetics and dynamics studies. Next, we use AWSEM to study protein-protein association. Our results indicate that the model not only can successfully predict the native dimeric interfaces but can also correctly reproduce the two and three state behavior of obligatory and nonobligatory dimers. We also find that both monomer geometry and specific non-bonded interactions play an important role in protein-protein association. Subsequently, we investigate protein folding under environmental fluctuations with a simple Go-like model. More specifically, we study the effect of an oscillating cellular environment on protein folding dynamics through modulating the strength of inter-residue interactions. The results show that, when occurring at some specific timescales, both deterministic and random fluctuations significantly accelerate the folding.