Premium
Prediction of hERG Potassium Channel Blocking Actions Using Combination of Classification and Regression Based Models: A Mixed Descriptors Approach
Author(s) -
Kar Supratik,
Roy Kunal
Publication year - 2012
Publication title -
molecular informatics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.481
H-Index - 68
eISSN - 1868-1751
pISSN - 1868-1743
DOI - 10.1002/minf.201200039
Subject(s) - herg , drugbank , quantitative structure–activity relationship , in silico , applicability domain , computer science , artificial intelligence , machine learning , linear discriminant analysis , data mining , computational biology , biological system , chemistry , potassium channel , pharmacology , biology , biochemistry , drug , gene , biophysics
A set of 242 compounds with diverse molecular structures and having different mechanisms of therapeutic actions was used to develop classification and regression based QSAR models for the identification of potential hERG channel blockers. The developed in silico models made it possible to obtain a quantitative interpretation of the structural information and physicochemical properties of the molecules for their hERG binding affinity along with exploration of the discriminant functions differentiating between lower and higher hERG blocking potency compounds by the classification approach. The developed models were rigorously validated internally as well as externally with the application of the principles of Organization for Economic Cooperation and Development (OECD) for the validation purpose. The test for domain of applicability was also carried out for checking reliability of the predictions. Pharmacological distribution diagrams (PDDs) were employed as a visualizing technique for the classification approach. Important fragments relevant to hERG binding affinity were identified through critical analysis and interpretation of the developed models. Finally, the developed models were implemented to screen hERG channel blocking properties for a huge number compounds of the DrugBank database. In silico prediction for hERG channel blocking potential for each of the DrugBank compounds is possible from the developed models.