Predicting Protein−Ligand Binding Affinities Using Novel Geometrical Descriptors and Machine-Learning Methods

Inspired by the concept of knowledge-based scoring functions, a new quantitative structure−activity relationship (QSAR) approach is introduced for scoring protein−ligand interactions. This approach considers that the strength of ligand binding is correlated with the nature of specific ligand/binding...

Full description

Saved in:
Bibliographic Details
Published inJournal of Chemical Information and Computer Sciences Vol. 44; no. 2; pp. 699 - 703
Main Authors Deng, Wei, Breneman, Curt, Embrechts, Mark J
Format Journal Article
LanguageEnglish
Published United States American Chemical Society 01.03.2004
Subjects
Algorithms
Artificial Intelligence
Databases, Protein
Ligands
Nonlinear Dynamics
Predictive Value of Tests
Protein Binding
Proteins - chemistry
Online AccessGet full text

Cover

More Information
Summary:Inspired by the concept of knowledge-based scoring functions, a new quantitative structure−activity relationship (QSAR) approach is introduced for scoring protein−ligand interactions. This approach considers that the strength of ligand binding is correlated with the nature of specific ligand/binding site atom pairs in a distance-dependent manner. In this technique, atom pair occurrence and distance-dependent atom pair features are used to generate an interaction score. Scoring and pattern recognition results obtained using Kernel PLS (partial least squares) modeling and a genetic algorithm-based feature selection method are discussed.
Bibliography:istex:2143FDF46B64326D967E08B75B604140980464F2
ark:/67375/TPS-0519H1PL-Q
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0095-2338
1549-960X
DOI:10.1021/ci034246+