ADMET evaluation in drug discovery. 12. Development of binary classification models for prediction of hERG potassium channel blockage

Sichao Wang, Youyong Li, Junmei Wang, Lei Chen, Liling Zhang, Huidong Yu, Tingjun Hou

Research output: Contribution to journalArticle

86 Citations (Scopus)

Abstract

Inhibition of the human ether-a-go-go related gene (hERG) potassium channel may result in QT interval prolongation, which causes severe cardiac side effects and is a major problem in clinical studies of drug candidates. The development of in silico tools to filter out potential hERG potassium channel blockers in early stages of the drug discovery process is of considerable interest. Here, a diverse set of 806 compounds with hERG inhibition data was assembled, and the binary hERG classification models using naive Bayesian classification and recursive partitioning (RP) techniques were established and evaluated. The naive Bayesian classifier based on molecular properties and the ECFP-8 fingerprints yielded 84.8% accuracy for the training set using the leave-one-out (LOO) cross-validation procedure and 85% accuracy for the test set of 120 molecules. For the two additional test sets, the model achieved 89.4% accuracy for the WOMBAT-PK test set, and 86.1% accuracy for the PubChem test set. The naive Bayesian classifiers gave better predictions than the RP classifiers. Moreover, the Bayesian classifier, employing molecular fingerprints, highlights the important structural fragments favorable or unfavorable for hERG potassium channel blockage, which offers extra valuable information for the design of compounds avoiding undesirable hERG activity.

Original languageEnglish (US)
Pages (from-to)996-1010
Number of pages15
JournalMolecular Pharmaceutics
Volume9
Issue number4
DOIs
StatePublished - Apr 2 2012

Fingerprint

Potassium Channels
Drug Discovery
Ether
Genes
Dermatoglyphics
Potassium Channel Blockers
Computer Simulation
Pharmaceutical Preparations

Keywords

  • ADMET
  • hERG
  • naive Bayesian classification
  • QSAR
  • recursive partitioning

ASJC Scopus subject areas

  • Pharmaceutical Science
  • Molecular Medicine
  • Drug Discovery

Cite this

ADMET evaluation in drug discovery. 12. Development of binary classification models for prediction of hERG potassium channel blockage. / Wang, Sichao; Li, Youyong; Wang, Junmei; Chen, Lei; Zhang, Liling; Yu, Huidong; Hou, Tingjun.

In: Molecular Pharmaceutics, Vol. 9, No. 4, 02.04.2012, p. 996-1010.

Research output: Contribution to journalArticle

Wang, Sichao ; Li, Youyong ; Wang, Junmei ; Chen, Lei ; Zhang, Liling ; Yu, Huidong ; Hou, Tingjun. / ADMET evaluation in drug discovery. 12. Development of binary classification models for prediction of hERG potassium channel blockage. In: Molecular Pharmaceutics. 2012 ; Vol. 9, No. 4. pp. 996-1010.
@article{5be0b8b1219942f9b247f2827079f674,
title = "ADMET evaluation in drug discovery. 12. Development of binary classification models for prediction of hERG potassium channel blockage",
abstract = "Inhibition of the human ether-a-go-go related gene (hERG) potassium channel may result in QT interval prolongation, which causes severe cardiac side effects and is a major problem in clinical studies of drug candidates. The development of in silico tools to filter out potential hERG potassium channel blockers in early stages of the drug discovery process is of considerable interest. Here, a diverse set of 806 compounds with hERG inhibition data was assembled, and the binary hERG classification models using naive Bayesian classification and recursive partitioning (RP) techniques were established and evaluated. The naive Bayesian classifier based on molecular properties and the ECFP-8 fingerprints yielded 84.8{\%} accuracy for the training set using the leave-one-out (LOO) cross-validation procedure and 85{\%} accuracy for the test set of 120 molecules. For the two additional test sets, the model achieved 89.4{\%} accuracy for the WOMBAT-PK test set, and 86.1{\%} accuracy for the PubChem test set. The naive Bayesian classifiers gave better predictions than the RP classifiers. Moreover, the Bayesian classifier, employing molecular fingerprints, highlights the important structural fragments favorable or unfavorable for hERG potassium channel blockage, which offers extra valuable information for the design of compounds avoiding undesirable hERG activity.",
keywords = "ADMET, hERG, naive Bayesian classification, QSAR, recursive partitioning",
author = "Sichao Wang and Youyong Li and Junmei Wang and Lei Chen and Liling Zhang and Huidong Yu and Tingjun Hou",
year = "2012",
month = "4",
day = "2",
doi = "10.1021/mp300023x",
language = "English (US)",
volume = "9",
pages = "996--1010",
journal = "Molecular Pharmaceutics",
issn = "1543-8384",
publisher = "American Chemical Society",
number = "4",

}

TY - JOUR

T1 - ADMET evaluation in drug discovery. 12. Development of binary classification models for prediction of hERG potassium channel blockage

AU - Wang, Sichao

AU - Li, Youyong

AU - Wang, Junmei

AU - Chen, Lei

AU - Zhang, Liling

AU - Yu, Huidong

AU - Hou, Tingjun

PY - 2012/4/2

Y1 - 2012/4/2

N2 - Inhibition of the human ether-a-go-go related gene (hERG) potassium channel may result in QT interval prolongation, which causes severe cardiac side effects and is a major problem in clinical studies of drug candidates. The development of in silico tools to filter out potential hERG potassium channel blockers in early stages of the drug discovery process is of considerable interest. Here, a diverse set of 806 compounds with hERG inhibition data was assembled, and the binary hERG classification models using naive Bayesian classification and recursive partitioning (RP) techniques were established and evaluated. The naive Bayesian classifier based on molecular properties and the ECFP-8 fingerprints yielded 84.8% accuracy for the training set using the leave-one-out (LOO) cross-validation procedure and 85% accuracy for the test set of 120 molecules. For the two additional test sets, the model achieved 89.4% accuracy for the WOMBAT-PK test set, and 86.1% accuracy for the PubChem test set. The naive Bayesian classifiers gave better predictions than the RP classifiers. Moreover, the Bayesian classifier, employing molecular fingerprints, highlights the important structural fragments favorable or unfavorable for hERG potassium channel blockage, which offers extra valuable information for the design of compounds avoiding undesirable hERG activity.

AB - Inhibition of the human ether-a-go-go related gene (hERG) potassium channel may result in QT interval prolongation, which causes severe cardiac side effects and is a major problem in clinical studies of drug candidates. The development of in silico tools to filter out potential hERG potassium channel blockers in early stages of the drug discovery process is of considerable interest. Here, a diverse set of 806 compounds with hERG inhibition data was assembled, and the binary hERG classification models using naive Bayesian classification and recursive partitioning (RP) techniques were established and evaluated. The naive Bayesian classifier based on molecular properties and the ECFP-8 fingerprints yielded 84.8% accuracy for the training set using the leave-one-out (LOO) cross-validation procedure and 85% accuracy for the test set of 120 molecules. For the two additional test sets, the model achieved 89.4% accuracy for the WOMBAT-PK test set, and 86.1% accuracy for the PubChem test set. The naive Bayesian classifiers gave better predictions than the RP classifiers. Moreover, the Bayesian classifier, employing molecular fingerprints, highlights the important structural fragments favorable or unfavorable for hERG potassium channel blockage, which offers extra valuable information for the design of compounds avoiding undesirable hERG activity.

KW - ADMET

KW - hERG

KW - naive Bayesian classification

KW - QSAR

KW - recursive partitioning

UR - http://www.scopus.com/inward/record.url?scp=84859345561&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859345561&partnerID=8YFLogxK

U2 - 10.1021/mp300023x

DO - 10.1021/mp300023x

M3 - Article

VL - 9

SP - 996

EP - 1010

JO - Molecular Pharmaceutics

JF - Molecular Pharmaceutics

SN - 1543-8384

IS - 4

ER -