Literature DB >> 26596162

Quantitative Assessment of Force Fields on Both Low-Energy Conformational Basins and Transition-State Regions of the (ϕ-ψ) Space.

Zhiwei Liu1, Bernd Ensing1, Preston B Moore1.   

Abstract

The free energy surfaces (FESs) of alanine dipeptide are studied to illustrate a new strategy to assess the performance of classical molecular mechanics force field on the full range of the (ϕ-ψ) conformational space. The FES is obtained from metadynamics simulations with five commonly used force fields and from ab initio density functional theory calculations in both gas phase and aqueous solution. The FESs obtained at the B3LYP/6-311+G(2d,p)//B3LYP/6-31G(d,p) level of theory are validated by comparison with previously reported MP2 and LMP2 results as well as with experimentally obtained probability distribution between the C5-β (or β-PPII) and αR states. A quantitative assessment is made for each force field in three conformational basins, LeRI (C5-β-C7eq), LeRII (β2-αR), and LeRIII(αL-C7ax-αD) as well as three transition-state regions linking the above conformational basins. The performance of each force field is evaluated in terms of the average free energy of each region in comparison with that of the ab initio results. We quantify how well a force field FES matches the ab initio FES through the calculation of the standard deviation of a free energy difference map between the two FESs. The results indicate that the performance varies largely from region to region or from force field to force field. Although not one force field is able to outperform all others in all conformational areas, the OPLSAA/L force field gives the best performance overall, followed by OPLSAA and AMBER03. For the three top performers, the average free energies differ from the corresponding ab initio values from within the error range (<0.4 kcal/mol) to ∼1.5 kcal/mol for the low-energy regions and up to ∼2.0 kcal/mol for the transition-state regions. The strategy presented and the results obtained here should be useful for improving the parametrization of force fields targeting both accuracy in the energies of conformers and the transition-state barriers.

Entities:  

Year:  2010        PMID: 26596162     DOI: 10.1021/ct100395n

Source DB:  PubMed          Journal:  J Chem Theory Comput        ISSN: 1549-9618            Impact factor:   6.006


  4 in total

1.  Electrostatic frequency shifts in amide I vibrational spectra: direct parameterization against experiment.

Authors:  Mike Reppert; Andrei Tokmakoff
Journal:  J Chem Phys       Date:  2013-04-07       Impact factor: 3.488

2.  Sequence-specific recognition of cancer drug-DNA adducts by HMGB1a repair protein.

Authors:  Robert M Elder; Arthi Jayaraman
Journal:  Biophys J       Date:  2012-05-15       Impact factor: 4.033

3.  A coupled two-dimensional main chain torsional potential for protein dynamics: generation and implementation.

Authors:  Yongxiu Li; Ya Gao; Xuqiang Zhang; Xingyu Wang; Lirong Mou; Lili Duan; Xiao He; Ye Mei; John Z H Zhang
Journal:  J Mol Model       Date:  2013-06-14       Impact factor: 1.810

4.  Correct folding of an α-helix and a β-hairpin using a polarized 2D torsional potential.

Authors:  Ya Gao; Yongxiu Li; Lirong Mou; Bingbing Lin; John Z H Zhang; Ye Mei
Journal:  Sci Rep       Date:  2015-06-03       Impact factor: 4.379

  4 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.