Identification of chemogenomic features from drug-target interaction networks using interpretable classifiers

Yasuo Tabei, Edouard Pauwels, Véronique Stoven, Kazuhiro Takemoto, Yoshihiro Yamanishi

Research output: Contribution to journalArticle

47 Citations (Scopus)

Abstract

Motivation: Drug effects are mainly caused by the interactions between drug molecules and their target proteins including primary targets and off-targets. Identification of the molecular mechanisms behind overall drug-target interactions is crucial in the drug design process. Results: We develop a classifier-based approach to identify chemogenomic features (the underlying associations between drug chemical substructures and protein domains) that are involved in drug-target interaction networks. We propose a novel algorithm for extracting informative chemogenomic features by using L1 regularized classifiers over the tensor product space of possible drug-target pairs. It is shown that the proposed method can extract a very limited number of chemogenomic features without loosing the performance of predicting drug-target interactions and the extracted features are biologically meaningful. The extracted substructure-domain association network enables us to suggest ligand chemical fragments specific for each protein domain and ligand core substructures important for a wide range of protein families.

Original languageEnglish
Article numberbts412
Pages (from-to)i487-i494
JournalBioinformatics
Volume28
Issue number18
DOIs
Publication statusPublished - Sep 1 2012

Fingerprint

Drug Interactions
Drugs
Classifiers
Classifier
Proteins
Target
Interaction
Pharmaceutical Preparations
Substructure
Protein
Ligands
Drug interactions
Drug Design
Tensors
Association reactions
Product Space
Molecules
Design Process
Tensor Product
Fragment

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

Identification of chemogenomic features from drug-target interaction networks using interpretable classifiers. / Tabei, Yasuo; Pauwels, Edouard; Stoven, Véronique; Takemoto, Kazuhiro; Yamanishi, Yoshihiro.

In: Bioinformatics, Vol. 28, No. 18, bts412, 01.09.2012, p. i487-i494.

Research output: Contribution to journalArticle

Tabei, Y, Pauwels, E, Stoven, V, Takemoto, K & Yamanishi, Y 2012, 'Identification of chemogenomic features from drug-target interaction networks using interpretable classifiers', Bioinformatics, vol. 28, no. 18, bts412, pp. i487-i494. https://doi.org/10.1093/bioinformatics/bts412
Tabei, Yasuo ; Pauwels, Edouard ; Stoven, Véronique ; Takemoto, Kazuhiro ; Yamanishi, Yoshihiro. / Identification of chemogenomic features from drug-target interaction networks using interpretable classifiers. In: Bioinformatics. 2012 ; Vol. 28, No. 18. pp. i487-i494.
@article{1433e5f298e545f29519cc04906f0789,
title = "Identification of chemogenomic features from drug-target interaction networks using interpretable classifiers",
abstract = "Motivation: Drug effects are mainly caused by the interactions between drug molecules and their target proteins including primary targets and off-targets. Identification of the molecular mechanisms behind overall drug-target interactions is crucial in the drug design process. Results: We develop a classifier-based approach to identify chemogenomic features (the underlying associations between drug chemical substructures and protein domains) that are involved in drug-target interaction networks. We propose a novel algorithm for extracting informative chemogenomic features by using L1 regularized classifiers over the tensor product space of possible drug-target pairs. It is shown that the proposed method can extract a very limited number of chemogenomic features without loosing the performance of predicting drug-target interactions and the extracted features are biologically meaningful. The extracted substructure-domain association network enables us to suggest ligand chemical fragments specific for each protein domain and ligand core substructures important for a wide range of protein families.",
author = "Yasuo Tabei and Edouard Pauwels and V{\'e}ronique Stoven and Kazuhiro Takemoto and Yoshihiro Yamanishi",
year = "2012",
month = "9",
day = "1",
doi = "10.1093/bioinformatics/bts412",
language = "English",
volume = "28",
pages = "i487--i494",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "18",

}

TY - JOUR

T1 - Identification of chemogenomic features from drug-target interaction networks using interpretable classifiers

AU - Tabei, Yasuo

AU - Pauwels, Edouard

AU - Stoven, Véronique

AU - Takemoto, Kazuhiro

AU - Yamanishi, Yoshihiro

PY - 2012/9/1

Y1 - 2012/9/1

N2 - Motivation: Drug effects are mainly caused by the interactions between drug molecules and their target proteins including primary targets and off-targets. Identification of the molecular mechanisms behind overall drug-target interactions is crucial in the drug design process. Results: We develop a classifier-based approach to identify chemogenomic features (the underlying associations between drug chemical substructures and protein domains) that are involved in drug-target interaction networks. We propose a novel algorithm for extracting informative chemogenomic features by using L1 regularized classifiers over the tensor product space of possible drug-target pairs. It is shown that the proposed method can extract a very limited number of chemogenomic features without loosing the performance of predicting drug-target interactions and the extracted features are biologically meaningful. The extracted substructure-domain association network enables us to suggest ligand chemical fragments specific for each protein domain and ligand core substructures important for a wide range of protein families.

AB - Motivation: Drug effects are mainly caused by the interactions between drug molecules and their target proteins including primary targets and off-targets. Identification of the molecular mechanisms behind overall drug-target interactions is crucial in the drug design process. Results: We develop a classifier-based approach to identify chemogenomic features (the underlying associations between drug chemical substructures and protein domains) that are involved in drug-target interaction networks. We propose a novel algorithm for extracting informative chemogenomic features by using L1 regularized classifiers over the tensor product space of possible drug-target pairs. It is shown that the proposed method can extract a very limited number of chemogenomic features without loosing the performance of predicting drug-target interactions and the extracted features are biologically meaningful. The extracted substructure-domain association network enables us to suggest ligand chemical fragments specific for each protein domain and ligand core substructures important for a wide range of protein families.

UR - http://www.scopus.com/inward/record.url?scp=84866467225&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866467225&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/bts412

DO - 10.1093/bioinformatics/bts412

M3 - Article

C2 - 22962471

AN - SCOPUS:84866467225

VL - 28

SP - i487-i494

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 18

M1 - bts412

ER -