FREE-ENERGY PERTURBATION CALCULATIONS ON PARALLEL COMPUTERS - DEMONSTRATIONS OF SCALABLE LINEAR SPEEDUP

Citation
Se. Debolt et al., FREE-ENERGY PERTURBATION CALCULATIONS ON PARALLEL COMPUTERS - DEMONSTRATIONS OF SCALABLE LINEAR SPEEDUP, Journal of computational chemistry, 15(3), 1994, pp. 351-373
Citations number
47
Categorie Soggetti
Chemistry
ISSN journal
01928651
Volume
15
Issue
3
Year of publication
1994
Pages
351 - 373
Database
ISI
SICI code
0192-8651(1994)15:3<351:FPCOPC>2.0.ZU;2-K
Abstract
A coarse-grain parallel implementation of the free energy perturbation (FEP) module of the AMBER molecular dynamics program is described and then demonstrated using five different molecular systems. The differe nce in the free energy of (aqueous) solvation is calculated for two mo novalent cations Delta Delta(aq)(Li+ --> Cs+), and for the zero-sum et hane-to-ethane' perturbation Delta Delta(aq)G (CH3-methyl-X --> X-meth yl--CH3), where X is a ghost methyl. The difference in binding free en ergy for a docked HIV-1 protease inhibitor into its ethylene mimetic i s examined by mutating its fifth peptide bond, Delta G(CO-NH --> CH=CH ). A potassium ion (K+) is driven outward from the center of mass of i onophore salinomycin (SAL(-)) in a potential of mean force calculation Delta G(MeOH)(SAL(-) . K+) carried out in methanol solvent. Parallel speedup obtained is linearly proportional to the number of parallel pr ocessors applied. Finally, the difference in free energy of solvation of phenol versus benzene, Delta Delta G(oct)(phenol --> benzene), is d etermined in water-saturated octanol and then expressed in terms of re lative partition coefficients, Delta log(P-o/w). Because no interproce ssor communication is required, this approach is scalable and applicab le in general for any parallel architecture or network of machines. FE P calculations run on the nCUBE/2 using 50 or 100 parallel processors were completed in clock times equivalent to or twice as fast as a Cray Y-MP. The difficulty of ensuring adequate system equilibrium when a g radual configurational reorientation follows the mutation of the Hamil tonian is discussed and analyzed. The results of a successful protocol for overcoming this equilibration problem are presented. The types of molecular perturbations for which this method is expected to perform most efficiently are described. (C) 1994 by John Wiley and Sons, Inc.