Benchmarking a wide range of chemical descriptors for drug-target interaction prediction using a chemogenomic approach

Ryusuke Sawada, Masaaki Kotera, Yoshihiro Yamanishi

研究成果: ジャーナルへの寄稿評論記事

21 引用 (Scopus)

抄録

(Figure Presented) The identification of drug-target interactions, or interactions between drug candidate compounds and target candidate proteins, is a crucial process in genomic drug discovery. In silico chemogenomic methods are recently recognized as a promising approach for genomewide scale prediction of drug-target interactions, but the prediction performance depends heavily on the descriptors and similarity measures of drugs and proteins. In this paper, we investigated the performance of various descriptors and similarity measures of drugs and proteins for the drug-target interaction prediction using a chemogenomic approach. We compared the prediction accuracy of 18 chemical descriptors of drugs (e.g., ECFP, FCFP,E-state, CDK, Klekota-Roth, MACCS, PubChem, Dragon, KCF-S, and graph kernels) and 4 descriptors of proteins (e.g., amino acid composition, domain profile, local sequence similarity, and string kernel) on about one hundred thousand drug-target interactions. We examined the combinatorial effects of drug descriptors and protein descriptors using the same benchmark data under several experimental conditions. Large-scale experiments showed that our proposed KCF-S descriptor worked the best in terms of prediction accuracy. The comparative results are expected to be useful for selecting chemical descriptors in various pharmaceutical applications.

元の言語英語
ページ(範囲)719-731
ページ数13
ジャーナルMolecular Informatics
33
発行部数11-12
DOI
出版物ステータス出版済み - 11 24 2014

Fingerprint

Benchmarking
Drug Interactions
Proteins
Pharmaceutical Preparations
Drug interactions
Drug Discovery
Computer Simulation
Drug products
Amino acids
Amino Acids
Chemical analysis
Experiments

All Science Journal Classification (ASJC) codes

  • Structural Biology
  • Molecular Medicine
  • Drug Discovery
  • Computer Science Applications
  • Organic Chemistry

これを引用

Benchmarking a wide range of chemical descriptors for drug-target interaction prediction using a chemogenomic approach. / Sawada, Ryusuke; Kotera, Masaaki; Yamanishi, Yoshihiro.

:: Molecular Informatics, 巻 33, 番号 11-12, 24.11.2014, p. 719-731.

研究成果: ジャーナルへの寄稿評論記事

Sawada, Ryusuke ; Kotera, Masaaki ; Yamanishi, Yoshihiro. / Benchmarking a wide range of chemical descriptors for drug-target interaction prediction using a chemogenomic approach. :: Molecular Informatics. 2014 ; 巻 33, 番号 11-12. pp. 719-731.
@article{57d1d4526591476ca09a28c29707911e,
title = "Benchmarking a wide range of chemical descriptors for drug-target interaction prediction using a chemogenomic approach",
abstract = "(Figure Presented) The identification of drug-target interactions, or interactions between drug candidate compounds and target candidate proteins, is a crucial process in genomic drug discovery. In silico chemogenomic methods are recently recognized as a promising approach for genomewide scale prediction of drug-target interactions, but the prediction performance depends heavily on the descriptors and similarity measures of drugs and proteins. In this paper, we investigated the performance of various descriptors and similarity measures of drugs and proteins for the drug-target interaction prediction using a chemogenomic approach. We compared the prediction accuracy of 18 chemical descriptors of drugs (e.g., ECFP, FCFP,E-state, CDK, Klekota-Roth, MACCS, PubChem, Dragon, KCF-S, and graph kernels) and 4 descriptors of proteins (e.g., amino acid composition, domain profile, local sequence similarity, and string kernel) on about one hundred thousand drug-target interactions. We examined the combinatorial effects of drug descriptors and protein descriptors using the same benchmark data under several experimental conditions. Large-scale experiments showed that our proposed KCF-S descriptor worked the best in terms of prediction accuracy. The comparative results are expected to be useful for selecting chemical descriptors in various pharmaceutical applications.",
author = "Ryusuke Sawada and Masaaki Kotera and Yoshihiro Yamanishi",
year = "2014",
month = "11",
day = "24",
doi = "10.1002/minf.201400066",
language = "English",
volume = "33",
pages = "719--731",
journal = "Molecular Informatics",
issn = "1868-1743",
publisher = "Wiley - VCH Verlag GmbH & CO. KGaA",
number = "11-12",

}

TY - JOUR

T1 - Benchmarking a wide range of chemical descriptors for drug-target interaction prediction using a chemogenomic approach

AU - Sawada, Ryusuke

AU - Kotera, Masaaki

AU - Yamanishi, Yoshihiro

PY - 2014/11/24

Y1 - 2014/11/24

N2 - (Figure Presented) The identification of drug-target interactions, or interactions between drug candidate compounds and target candidate proteins, is a crucial process in genomic drug discovery. In silico chemogenomic methods are recently recognized as a promising approach for genomewide scale prediction of drug-target interactions, but the prediction performance depends heavily on the descriptors and similarity measures of drugs and proteins. In this paper, we investigated the performance of various descriptors and similarity measures of drugs and proteins for the drug-target interaction prediction using a chemogenomic approach. We compared the prediction accuracy of 18 chemical descriptors of drugs (e.g., ECFP, FCFP,E-state, CDK, Klekota-Roth, MACCS, PubChem, Dragon, KCF-S, and graph kernels) and 4 descriptors of proteins (e.g., amino acid composition, domain profile, local sequence similarity, and string kernel) on about one hundred thousand drug-target interactions. We examined the combinatorial effects of drug descriptors and protein descriptors using the same benchmark data under several experimental conditions. Large-scale experiments showed that our proposed KCF-S descriptor worked the best in terms of prediction accuracy. The comparative results are expected to be useful for selecting chemical descriptors in various pharmaceutical applications.

AB - (Figure Presented) The identification of drug-target interactions, or interactions between drug candidate compounds and target candidate proteins, is a crucial process in genomic drug discovery. In silico chemogenomic methods are recently recognized as a promising approach for genomewide scale prediction of drug-target interactions, but the prediction performance depends heavily on the descriptors and similarity measures of drugs and proteins. In this paper, we investigated the performance of various descriptors and similarity measures of drugs and proteins for the drug-target interaction prediction using a chemogenomic approach. We compared the prediction accuracy of 18 chemical descriptors of drugs (e.g., ECFP, FCFP,E-state, CDK, Klekota-Roth, MACCS, PubChem, Dragon, KCF-S, and graph kernels) and 4 descriptors of proteins (e.g., amino acid composition, domain profile, local sequence similarity, and string kernel) on about one hundred thousand drug-target interactions. We examined the combinatorial effects of drug descriptors and protein descriptors using the same benchmark data under several experimental conditions. Large-scale experiments showed that our proposed KCF-S descriptor worked the best in terms of prediction accuracy. The comparative results are expected to be useful for selecting chemical descriptors in various pharmaceutical applications.

UR - http://www.scopus.com/inward/record.url?scp=84915753460&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84915753460&partnerID=8YFLogxK

U2 - 10.1002/minf.201400066

DO - 10.1002/minf.201400066

M3 - Review article

AN - SCOPUS:84915753460

VL - 33

SP - 719

EP - 731

JO - Molecular Informatics

JF - Molecular Informatics

SN - 1868-1743

IS - 11-12

ER -