In silico prediction of pregnane X receptor activators by machine learning approaches

C. Y. Ung, Hu Li, C. W. Yap, Y. Z. Chen

Research output: Contribution to journalArticle

70 Citations (Scopus)

Abstract

Pregnane X receptor (PXR) regulates drug metabolism and is involved in drug-drug interactions. Prediction of PXR activators is important for evaluating drug metabolism and toxicity. Computational pharmacophore and quantitative structure-activity relationship models have been developed for predicting PXR activators. Because of the structural diversity of PXR activators, more efforts are needed for exploring methods applicable to a broader spectrum of compounds. We explored three machine learning methods (MLMs) for predicting PXR activators, which were trained and tested by using significantly higher number of compounds, 128 PXR activators (98 human) and 77 PXR non-activators, than those of previous studies. The recursive feature-selection method was used to select molecular descriptors relevant to PXR activator prediction, which are consistent with conclusions from other computational and structural studies. In a 10-fold cross-validation test, our MLM systems correctly predicted 81.2 to 84.0% of PXR activators, 80.8 to 85.0% of hPXR activators, 61.2 to 70.3% of PXR nonactivators, and 67.7 to 73.6% of hPXR nonactivators. Our systems also correctly predicted 73.3 to 86.7% of 15 newly published hPXR activators. MLMs seem to be useful for predicting PXR activators and for providing clues to physicochemical features of PXR activation.

Original languageEnglish (US)
Pages (from-to)158-168
Number of pages11
JournalMolecular Pharmacology
Volume71
Issue number1
DOIs
StatePublished - 2007
Externally publishedYes

Fingerprint

Computer Simulation
pregnane X receptor
Machine Learning
Quantitative Structure-Activity Relationship
Drug-Related Side Effects and Adverse Reactions
Drug Interactions
Pharmaceutical Preparations

ASJC Scopus subject areas

  • Pharmacology

Cite this

In silico prediction of pregnane X receptor activators by machine learning approaches. / Ung, C. Y.; Li, Hu; Yap, C. W.; Chen, Y. Z.

In: Molecular Pharmacology, Vol. 71, No. 1, 2007, p. 158-168.

Research output: Contribution to journalArticle

Ung, C. Y. ; Li, Hu ; Yap, C. W. ; Chen, Y. Z. / In silico prediction of pregnane X receptor activators by machine learning approaches. In: Molecular Pharmacology. 2007 ; Vol. 71, No. 1. pp. 158-168.
@article{262d8e6ca5b64065b28f92de9aba31c2,
title = "In silico prediction of pregnane X receptor activators by machine learning approaches",
abstract = "Pregnane X receptor (PXR) regulates drug metabolism and is involved in drug-drug interactions. Prediction of PXR activators is important for evaluating drug metabolism and toxicity. Computational pharmacophore and quantitative structure-activity relationship models have been developed for predicting PXR activators. Because of the structural diversity of PXR activators, more efforts are needed for exploring methods applicable to a broader spectrum of compounds. We explored three machine learning methods (MLMs) for predicting PXR activators, which were trained and tested by using significantly higher number of compounds, 128 PXR activators (98 human) and 77 PXR non-activators, than those of previous studies. The recursive feature-selection method was used to select molecular descriptors relevant to PXR activator prediction, which are consistent with conclusions from other computational and structural studies. In a 10-fold cross-validation test, our MLM systems correctly predicted 81.2 to 84.0{\%} of PXR activators, 80.8 to 85.0{\%} of hPXR activators, 61.2 to 70.3{\%} of PXR nonactivators, and 67.7 to 73.6{\%} of hPXR nonactivators. Our systems also correctly predicted 73.3 to 86.7{\%} of 15 newly published hPXR activators. MLMs seem to be useful for predicting PXR activators and for providing clues to physicochemical features of PXR activation.",
author = "Ung, {C. Y.} and Hu Li and Yap, {C. W.} and Chen, {Y. Z.}",
year = "2007",
doi = "10.1124/mol.106.027623",
language = "English (US)",
volume = "71",
pages = "158--168",
journal = "Molecular Pharmacology",
issn = "0026-895X",
publisher = "American Society for Pharmacology and Experimental Therapeutics",
number = "1",

}

TY - JOUR

T1 - In silico prediction of pregnane X receptor activators by machine learning approaches

AU - Ung, C. Y.

AU - Li, Hu

AU - Yap, C. W.

AU - Chen, Y. Z.

PY - 2007

Y1 - 2007

N2 - Pregnane X receptor (PXR) regulates drug metabolism and is involved in drug-drug interactions. Prediction of PXR activators is important for evaluating drug metabolism and toxicity. Computational pharmacophore and quantitative structure-activity relationship models have been developed for predicting PXR activators. Because of the structural diversity of PXR activators, more efforts are needed for exploring methods applicable to a broader spectrum of compounds. We explored three machine learning methods (MLMs) for predicting PXR activators, which were trained and tested by using significantly higher number of compounds, 128 PXR activators (98 human) and 77 PXR non-activators, than those of previous studies. The recursive feature-selection method was used to select molecular descriptors relevant to PXR activator prediction, which are consistent with conclusions from other computational and structural studies. In a 10-fold cross-validation test, our MLM systems correctly predicted 81.2 to 84.0% of PXR activators, 80.8 to 85.0% of hPXR activators, 61.2 to 70.3% of PXR nonactivators, and 67.7 to 73.6% of hPXR nonactivators. Our systems also correctly predicted 73.3 to 86.7% of 15 newly published hPXR activators. MLMs seem to be useful for predicting PXR activators and for providing clues to physicochemical features of PXR activation.

AB - Pregnane X receptor (PXR) regulates drug metabolism and is involved in drug-drug interactions. Prediction of PXR activators is important for evaluating drug metabolism and toxicity. Computational pharmacophore and quantitative structure-activity relationship models have been developed for predicting PXR activators. Because of the structural diversity of PXR activators, more efforts are needed for exploring methods applicable to a broader spectrum of compounds. We explored three machine learning methods (MLMs) for predicting PXR activators, which were trained and tested by using significantly higher number of compounds, 128 PXR activators (98 human) and 77 PXR non-activators, than those of previous studies. The recursive feature-selection method was used to select molecular descriptors relevant to PXR activator prediction, which are consistent with conclusions from other computational and structural studies. In a 10-fold cross-validation test, our MLM systems correctly predicted 81.2 to 84.0% of PXR activators, 80.8 to 85.0% of hPXR activators, 61.2 to 70.3% of PXR nonactivators, and 67.7 to 73.6% of hPXR nonactivators. Our systems also correctly predicted 73.3 to 86.7% of 15 newly published hPXR activators. MLMs seem to be useful for predicting PXR activators and for providing clues to physicochemical features of PXR activation.

UR - http://www.scopus.com/inward/record.url?scp=33845870927&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845870927&partnerID=8YFLogxK

U2 - 10.1124/mol.106.027623

DO - 10.1124/mol.106.027623

M3 - Article

C2 - 17003167

AN - SCOPUS:33845870927

VL - 71

SP - 158

EP - 168

JO - Molecular Pharmacology

JF - Molecular Pharmacology

SN - 0026-895X

IS - 1

ER -