Empirical aqueous solvation models based on accessible surface areas   with implicit electrostatics

doi:10.1021/jp025595u

CORC > 北京大学 > 化学与分子工程学院

	Empirical aqueous solvation models based on accessible surface areas with implicit electrostatics
	Hou, TJ ; Qiao, XB ; Zhang, W ; Xu, XJ
刊名	journal of physical chemistry b
	2002
关键词	GROWTH-FACTOR RECEPTOR TYROSINE KINASE INHIBITORS FREE-ENERGIES GENETIC ALGORITHM ORGANIC-MOLECULES AUTOMATED DOCKING FORCE-FIELD PROTEINS PEPTIDES SOLVENT
DOI	10.1021/jp025595u
英文摘要	In the current work, an empirical solvation model based on accessible surface areas was reported, which can be used to predict the solvation free energies for both organic and biological molecules very fast. This solvation model is based on atom-weighted solvent accessible surface area (SAWSA). The parameterization procedure for different kinds of atoms was performed as follows: first, the atoms in a molecule were defined to different atom types based on SMARTS language; then the solvent accessible surface area for each atom (or charged group) was calculated; finally, 4 genetic algorithm (GA) was applied to optimize the solvation parameters for different atom types in order to reproduce the experimental solvation free energies. The derived model possessed promising predictive ability as indicated by the good statistical significance and good prediction on the external test set. Using the solvation model based on all 377 neutral molecules, we have achieved an average unsigned error of 0:51 kcal/mol and standard deviation of 0.46 kcal/mol, which was better than the model proposed by Wang et al. The solvation model developed in the current work was applied to predict the solvation free energies of small organic molecules and proteins. For the,51 small organic molecules, the SAWSA model could give consistent results with the AM1/SM2.1 model in addition to several molecules with large conjugate systems. Moreover, the predictions from SAWSA were much better than those from SM5.0R, a solvation model based on geometry-dependent atom surface tensions. For the IS proteins randomly selected from the Brookhaven PDB database, the solvation free energies predicted by the SAWSA model showed high linear correlation (r = 0.99) than those predicted by PBSA, which were much better than those given by the Ooi model and the Vial model. Finally, we have successfully applied this model to predict the relative binding free energies for four binding modes of EGFR/quinazoline. The most favorable binding mode identified by MM-PBSA could also be correctly recognized by MM-SAWSA. The relative solvation free energies calculated by SAWSA show obvious correlation with those calculated by PBSA. The SAWSA should have potential applications in QSAR, molecular docking, protein folding, free energy calculations, and so forth.; Chemistry, Physical; SCI(E); EI; 29; ARTICLE; 43; 11295-11304; 106
语种	英语
内容类型	期刊论文
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/208388]
专题	化学与分子工程学院
推荐引用方式 GB/T 7714	Hou, TJ,Qiao, XB,Zhang, W,et al. Empirical aqueous solvation models based on accessible surface areas with implicit electrostatics[J]. journal of physical chemistry b,2002.
APA	Hou, TJ,Qiao, XB,Zhang, W,&Xu, XJ.(2002).Empirical aqueous solvation models based on accessible surface areas with implicit electrostatics.journal of physical chemistry b.
MLA	Hou, TJ,et al."Empirical aqueous solvation models based on accessible surface areas with implicit electrostatics".journal of physical chemistry b (2002).