Coarse-grained（CG） simulations can more efficiently study large conformational changes of biological polymers but usually lose accuracies in the details. Lots of different hybrid models involving multiple different ...Coarse-grained（CG） simulations can more efficiently study large conformational changes of biological polymers but usually lose accuracies in the details. Lots of different hybrid models involving multiple different resolutions have been developed to overcome the difficulty. Here we propose a novel effective hybrid CG（hyCG） approach which mixes the fine-grained interaction and its average in CG space to form a more smoothing potential energy surface. The hyCG approximately reproduces the potential of mean force in the CG space, and multiple mixed potentials can be further combined together to form a single effective force field for achieving both high efficiency and high accuracy. We illustrate the hyCG method in Trp-cage and Villin headpiece proteins to exhibit the folding of proteins. The topology of the folding landscape and thus the folding paths are preserved, while the folding is boosted nearly one order of magnitude faster. It indicates that the hyCG approach could be applied as an efficient force field in proteins.展开更多
基金Project supported by the National Basic Research Program of China(Grant No.2013CB932803)the National Natural Science Foundation of China(Grant No.11574310)+2 种基金the Joint NSFC-ISF Research Programjointly funded by the National Natural Science Foundation of Chinathe Israel Science Foundation(Grant No.51561145002)
文摘Coarse-grained（CG） simulations can more efficiently study large conformational changes of biological polymers but usually lose accuracies in the details. Lots of different hybrid models involving multiple different resolutions have been developed to overcome the difficulty. Here we propose a novel effective hybrid CG（hyCG） approach which mixes the fine-grained interaction and its average in CG space to form a more smoothing potential energy surface. The hyCG approximately reproduces the potential of mean force in the CG space, and multiple mixed potentials can be further combined together to form a single effective force field for achieving both high efficiency and high accuracy. We illustrate the hyCG method in Trp-cage and Villin headpiece proteins to exhibit the folding of proteins. The topology of the folding landscape and thus the folding paths are preserved, while the folding is boosted nearly one order of magnitude faster. It indicates that the hyCG approach could be applied as an efficient force field in proteins.
