Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions
In this paper, a coefficient adaptive scoring method of molecular docking is presented to improve the docking accuracy with multiple available scoring functions. Based on force-field scoring function, we considered hydrophobic and deformation as well in the proposed method, Instead of simple combina...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Bulgarian Academy of Sciences
2014-03-01
|
Series: | International Journal Bioautomation |
Subjects: | |
Online Access: | http://www.biomed.bas.bg/bioautomation/2014/vol_18.1/files/18.1_01.pdf |
id |
doaj-3f769172b2b443deb9964de9303601a4 |
---|---|
record_format |
Article |
spelling |
doaj-3f769172b2b443deb9964de9303601a42020-11-25T03:22:00ZengBulgarian Academy of SciencesInternational Journal Bioautomation1314-19021314-23212014-03-01181514Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring FunctionsZhengfu Li0Xicheng WangKeqiu LiJunfeng GuLing KangSchool of Computer Science and Technology, Dalian University of Technology, 2 Linggong Road, Dalian, P.R. China, 116024In this paper, a coefficient adaptive scoring method of molecular docking is presented to improve the docking accuracy with multiple available scoring functions. Based on force-field scoring function, we considered hydrophobic and deformation as well in the proposed method, Instead of simple combination with fixed weights, coefficients of each factor are adaptive in searching procedure. In order to improve the docking accuracy and stability, knowledge-based scoring function is used as another scoring factor. Genetic algorithm with the multi-population evolution and entropy-based searching technique with narrowing down space is used to solve the optimization model for molecular docking. To evaluate the method, we carried out a numerical experiment with 134 protein- ligand complexes of the publicly available GOLD test set. The results validated that it improved the docking accuracy over the individual force-field scoring. In addition, analyses were given to show the disadvantage of individual scoring model. Through the comparison with other popular docking software, the proposed method showed higher accuracy. Among more than 77% of the complexes, the docked results were within 1.0 Å according to Root- Mean-Square Deviation (RMSD) of the X-ray structure. The average computing time obtained here is 563.9 s.http://www.biomed.bas.bg/bioautomation/2014/vol_18.1/files/18.1_01.pdfGenetic algorithmsCoefficient adaptiveMolecular dockingScoring functionOptimization |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Zhengfu Li Xicheng Wang Keqiu Li Junfeng Gu Ling Kang |
spellingShingle |
Zhengfu Li Xicheng Wang Keqiu Li Junfeng Gu Ling Kang Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions International Journal Bioautomation Genetic algorithms Coefficient adaptive Molecular docking Scoring function Optimization |
author_facet |
Zhengfu Li Xicheng Wang Keqiu Li Junfeng Gu Ling Kang |
author_sort |
Zhengfu Li |
title |
Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions |
title_short |
Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions |
title_full |
Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions |
title_fullStr |
Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions |
title_full_unstemmed |
Molecular Docking Improvement: Coefficient Adaptive Genetic Algorithms for Multiple Scoring Functions |
title_sort |
molecular docking improvement: coefficient adaptive genetic algorithms for multiple scoring functions |
publisher |
Bulgarian Academy of Sciences |
series |
International Journal Bioautomation |
issn |
1314-1902 1314-2321 |
publishDate |
2014-03-01 |
description |
In this paper, a coefficient adaptive scoring method of molecular docking is presented to improve the docking accuracy with multiple available scoring functions. Based on force-field scoring function, we considered hydrophobic and deformation as well in the proposed method, Instead of simple combination with fixed weights, coefficients of each factor are adaptive in searching procedure. In order to improve the docking accuracy and stability, knowledge-based scoring function is used as another scoring factor. Genetic algorithm with the multi-population evolution and entropy-based searching technique with narrowing down space is used to solve the optimization model for molecular docking. To evaluate the method, we carried out a numerical experiment with 134 protein- ligand complexes of the publicly available GOLD test set. The results validated that it improved the docking accuracy over the individual force-field scoring. In addition, analyses were given to show the disadvantage of individual scoring model. Through the comparison with other popular docking software, the proposed method showed higher accuracy. Among more than 77% of the complexes, the docked results were within 1.0 Å according to Root- Mean-Square Deviation (RMSD) of the X-ray structure. The average computing time obtained here is 563.9 s. |
topic |
Genetic algorithms Coefficient adaptive Molecular docking Scoring function Optimization |
url |
http://www.biomed.bas.bg/bioautomation/2014/vol_18.1/files/18.1_01.pdf |
work_keys_str_mv |
AT zhengfuli moleculardockingimprovementcoefficientadaptivegeneticalgorithmsformultiplescoringfunctions AT xichengwang moleculardockingimprovementcoefficientadaptivegeneticalgorithmsformultiplescoringfunctions AT keqiuli moleculardockingimprovementcoefficientadaptivegeneticalgorithmsformultiplescoringfunctions AT junfenggu moleculardockingimprovementcoefficientadaptivegeneticalgorithmsformultiplescoringfunctions AT lingkang moleculardockingimprovementcoefficientadaptivegeneticalgorithmsformultiplescoringfunctions |
_version_ |
1724611849439150080 |