Abstract
AbstractThe interpretation of single-molecule experiments is frequently aided by computational modeling of biomolecular dynamics. The growth of computing power and ongoing validation of computational models suggest that it soon may be possible to replace some experiments out-right with computational mimics. Here we offer a blueprint for performing single-molecule studies in silico using a DNA binding protein as a test bed. We demonstrate how atomistic simulations, typically limited to sub-millisecond durations and zeptoliter volumes, can guide development of a coarse-grained model for use in simulations that mimic experimental assays. We show that, after initially correcting excess attraction between the DNA and protein, qualitative consistency between several experiments and their computational equivalents is achieved, while additionally providing a detailed portrait of the underlying mechanics. Finally the model is used to simulate the trombone loop of a replication fork, a large complex of proteins and DNA.
Publisher
Cold Spring Harbor Laboratory