Improved Prediction of Protein Ligand-Binding Sites Using Random Forests

Title: Improved Prediction of Protein Ligand-Binding Sites Using Random Forests

Volume: 18 Issue: 12

Author(s): Zhijun Qiu and Xicheng Wang

Affiliation:

Keywords: ligand-binding site prediction, patch-based residue characterization, random forests, Q-SiteFinder, SCREEN, Morita's method, X-ray crystallography, NMR, hydrophobicity, pocket depth, jackknife test, ASA, Feature vector for residue, PSAIA, Solvation energyligand-binding site prediction, patch-based residue characterization, random forests, Q-SiteFinder, SCREEN, Morita's method, X-ray crystallography, NMR, hydrophobicity, pocket depth, jackknife test, ASA, Feature vector for residue, PSAIA, Solvation energy

Abstract: This article describes a novel method for predicting ligand-binding sites of proteins. This method uses only 8 structural properties as input vector to train 9 random forest classifiers which are combined to predict binding residues. These predicted binding residues are then clustered into some predicted ligand-binding sites. According to our measurement criterion, this method achieved a success rate of 0.914 in the bound state dataset and 0.800 in the unbound state dataset, which are better than three other methods: Q-SiteFinder, SCREEN and Moritas method. It indicates that the proposed method here is successful for predicting ligand-binding sites.

Cite this article as:

Qiu Zhijun and Wang Xicheng, Improved Prediction of Protein Ligand-Binding Sites Using Random Forests, Protein & Peptide Letters 2011; 18 (12) . https://dx.doi.org/10.2174/092986611797642788