AUTHOR=Liu Liu , Hu Xiuzhen , Feng Zhenxing , Wang Shan , Sun Kai , Xu Shuang 

TITLE=Recognizing Ion Ligand–Binding Residues by Random Forest Algorithm Based on Optimized Dihedral Angle

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 8 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2020.00493

DOI=10.3389/fbioe.2020.00493

ISSN=2296-4185

ABSTRACT=The prediction of ion ligand binding residues on proteins helps to understand the specific functions of proteins in life processes. At present, it is a challenge work to predict binding residues of ion ligands on proteins in the studies of protein functions. In this paper, we constructed datasets with 14 ion ligand binding residues, including 4 acid radical ion ligands and 10 metal ion ligands. Based on the amino acid sequence information, we selected the composition and position conservation information of amino acids, the predicted structural information and physicochemical properties of amino acids as basic feature parameters. We then made a statistical analysis and reclassification for dihedral angle, and proposed new methods in the extraction of feature parameters. The methods mainly included the extraction of polarization charge and hydrophilic-hydrophobic information of amino acids by using information entropy, and the extraction of position conservation information by using position weight matrices. In the prediction model, we got better prediction results than previous works. With the 5-fold cross-validation, the Matthew’s correlation coefficient and accuracy of 10 metal ion ligand binding residues were higher than 0.61 and 80%, respectively; the above values of the 4 acid radical ion ligand binding residues were higher than 0.54 and 76%, respectively. Further, we classified and combined the phi and psi angles, and obtained the optimal prediction model for each ion ligand binding residues. Finally, we performed an independent test on the prediction model, the well results indicate the practicability for our proposed method.