Development of classification model for hERG ion channel inhibitors using SVM method

Sinmoon Gang, Hanjo Kim, Wonseok Oh, Sunyoung Kim, Kyoung Tai No, Ky Youb Nam

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.

Original languageEnglish
Pages (from-to)653-662
Number of pages10
JournalJournal of the Korean Chemical Society
Volume53
Issue number6
DOIs
Publication statusPublished - 2009 Dec 20

Fingerprint

Ion Channels
Support vector machines
Ions
Metabolism
Toxicity
Feature extraction
Testing
Pharmaceutical Preparations

All Science Journal Classification (ASJC) codes

  • Chemistry (miscellaneous)
  • Chemical Engineering (miscellaneous)

Cite this

Gang, Sinmoon ; Kim, Hanjo ; Oh, Wonseok ; Kim, Sunyoung ; No, Kyoung Tai ; Nam, Ky Youb. / Development of classification model for hERG ion channel inhibitors using SVM method. In: Journal of the Korean Chemical Society. 2009 ; Vol. 53, No. 6. pp. 653-662.
@article{35a136d280a64c0d9db53f1a36a15269,
title = "Development of classification model for hERG ion channel inhibitors using SVM method",
abstract = "Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.",
author = "Sinmoon Gang and Hanjo Kim and Wonseok Oh and Sunyoung Kim and No, {Kyoung Tai} and Nam, {Ky Youb}",
year = "2009",
month = "12",
day = "20",
doi = "10.5012/jkcs.2009.53.6.653",
language = "English",
volume = "53",
pages = "653--662",
journal = "Journal of the Korean Chemical Society",
issn = "1017-2548",
publisher = "Korean Chemical Society",
number = "6",

}

Development of classification model for hERG ion channel inhibitors using SVM method. / Gang, Sinmoon; Kim, Hanjo; Oh, Wonseok; Kim, Sunyoung; No, Kyoung Tai; Nam, Ky Youb.

In: Journal of the Korean Chemical Society, Vol. 53, No. 6, 20.12.2009, p. 653-662.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development of classification model for hERG ion channel inhibitors using SVM method

AU - Gang, Sinmoon

AU - Kim, Hanjo

AU - Oh, Wonseok

AU - Kim, Sunyoung

AU - No, Kyoung Tai

AU - Nam, Ky Youb

PY - 2009/12/20

Y1 - 2009/12/20

N2 - Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.

AB - Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.

UR - http://www.scopus.com/inward/record.url?scp=73549084973&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=73549084973&partnerID=8YFLogxK

U2 - 10.5012/jkcs.2009.53.6.653

DO - 10.5012/jkcs.2009.53.6.653

M3 - Article

VL - 53

SP - 653

EP - 662

JO - Journal of the Korean Chemical Society

JF - Journal of the Korean Chemical Society

SN - 1017-2548

IS - 6

ER -