A Two-Step Target Binding and Selectivity Support Vector Machines Approach for Virtual Screening of Dopamine Receptor Subtype-Selective Ligands

Yuyang Jiang,Yuzong Chen,Xiaona Wei,Jingxian Zhang,Bucong Han,Chunyan Tan,Arto Urtti

doi:10.1371/journal.pone.0039076

Yuyang Jiang, Yuzong Chen + Show 5 more

Open Access

https://doi.org/10.1371/journal.pone.0039076

Copy DOI

Abstract

Target selective drugs, such as dopamine receptor (DR) subtype selective ligands, are developed for enhanced therapeutics and reduced side effects. In silico methods have been explored for searching DR selective ligands, but encountered difficulties associated with high subtype similarity and ligand structural diversity. Machine learning methods have shown promising potential in searching target selective compounds. Their target selective capability can be further enhanced. In this work, we introduced a new two-step support vector machines target-binding and selectivity screening method for searching DR subtype-selective ligands, which was tested together with three previously-used machine learning methods for searching D1, D2, D3 and D4 selective ligands. It correctly identified 50.6%–88.0% of the 21–408 subtype selective and 71.7%–81.0% of the 39–147 multi-subtype ligands. Its subtype selective ligand identification rates are significantly better than, and its multi-subtype ligand identification rates are comparable to the best rates of the previously used methods. Our method produced low false-hit rates in screening 13.56 M PubChem, 168,016 MDDR and 657,736 ChEMBLdb compounds. Molecular features important for subtype selectivity were extracted by using the recursive feature elimination feature selection method. These features are consistent with literature-reported features. Our method showed similar performance in searching estrogen receptor subtype selective ligands. Our study demonstrated the usefulness of the two-step target binding and selectivity screening method in searching subtype selective ligands from large compound libraries.

Highlights

Drugs that selectively modulate protein subtypes are highly useful for achieving therapeutic efficacies at reduced side effects [1,2,3,4]
Our new method 2SBR-SVM was tested together with three previously-used methods Combi-SVM [18] and two methods in the Mulan software package [30]: the multi-label knearest-neighbor (ML-kNN) [28,31] and Random k-labelset Decision Tree (RAkEL-DT) [32,33] methods. The purpose of these tests was to evaluate the performance of the previously used methods, and to determine to what extent our new method can improve the performance in selecting dopamine subtype selective ligands
Virtual screening methods have been increasingly explored for facilitating the discovery of target selective drugs for enhanced therapeutics and reduced side effects

Summary

Introduction

Drugs that selectively modulate protein subtypes are highly useful for achieving therapeutic efficacies at reduced side effects [1,2,3,4]. For some targets such as dopamine receptors, all of the approved drugs are subtype non-selective, and this non-selectivity directly contributes to their observed side effects and adversely affects their application potential [4]. Ligand binding selectivity to these subtypes is both determined by the structural and physicochemical features of the conserved and nonconserved residues [9]. D2/D4 selectivity has been suggested to be determined by mutated residues within the second, third, and seventh membrane-spanning segments [9]

Methods

Results

Conclusion