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.Availability: Softwares are available at the supplemental website.Contact: yamanishi@bioreg.kyushu-u.ac.jpSupplementary Information: Datasets and all results are available at http://cbio.ensmp.fr/~yyamanishi/l1binary/ .
Highlights
Drug phenotypic effects are caused by the interactions between drug molecules and their target proteins including their primary targets and off-targets (Blagg, 2006; Whitebread et al, 2005)
We develop a classifier-based approach to identify chemogenomic features which are strongly involved in drug–target interaction networks
Each chemogenomic feature consists of a chemical substructure and a protein domain which are suspected of being associated with each other in terms of drug–target interactions
Summary
Drug phenotypic effects are caused by the interactions between drug molecules and their target proteins including their primary targets and off-targets (Blagg, 2006; Whitebread et al, 2005). Chemogenomics is an emerging research area that attempts to associate the chemical space of possible ligands with the genomic space of possible proteins (Dobson, 2004; Kanehisa et al, 2006; Stockwell, 2000). Following this principle, several statistical methods have been proposed to predict drug–target or ligand– protein interactions on a large scale. (Faulon et al, 2008; Jacob and Vert, 2008; Keiser et al, 2009; Li et al, 2011; Yabuuchi et al, 2011; Yamanishi et al, 2008; Yang et al, 2009) These methods are purely predictive and do not provide any further understanding of molecular mechanisms behind ligand–protein interactions. Most previous works have been performed from the viewpoint of either chemical substructures or protein functional sites
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have