Abstract
Accurate approximations are introduced for the evaluation of inverse interparticle distances such that square root or division operations are not required in the computation of interparticle interactions and forces. These generally applicable approximations are illustrated by incorporation into the protein simulation package TINKER (Pappu, R. V.; Hart, R. K.; Ponder, J. W. J Phys Chem B 1998, 102, 9725; Ponder, J. W.; Richards, F. M. J Comput Chem 1987, 8, 1016; Dudek, M. J.; Ramnarayan, K.; Ponder, J. W. J Comput Chem 1998, 19, 548) along with several other speed enhancement strategies. With these modifications, implicit solvent Langevin dynamics simulations of proteins are performed factors of 4.6 times faster than the modified open source distributed program. Programming speedups are obtained by extensive vectorization, simplification of the inner loop to avoid IF statements, and by using lookup tables for the distance dependent โdielectric constantโ in implicit solvent models. Benchmarks are provided for the allโatom, implicit solvent dynamics of Metโenkephalin, the villin headpiece, the B1 domain of proteinโG, and barnase. We also discuss the more general applicability of the approximation methods to explicit solvent simulations and of lookโup tables for other implicit solvent models such as the generalized Born models. ยฉ 2005 Wiley Periodicals, Inc. J Comput Chem 26: 691โ698, 2005