Amino Acid Features for Prediction of Protein-Protein Interface Residues with Support Vector Machines

Abstract

Abstract. Knowledge of protein-protein interaction sites is vital to determine proteins' function and involvement in different pathways. Support Vector Machines (SVM) have been proposed over the recent years to predict protein-protein interface residues, primarily based on single amino acid sequence inputs. We investigate the features of amino acids that can be best used with SVM for predicting residues at proteinprotein interfaces. The optimal feature set was derived from investigation into features such as amino acid composition, hydrophobic characters of amino acids, secondary structure propensity of amino acids, accessible surface areas, and evolutionary information generated by PSI-BLAST profiles. Using a backward elimination procedure, amino acid composition, accessible surface areas, and evolutionary information generated by PSI-BLAST profiles gave the best performance. The present approach achieved overall prediction accuracy of 74.2% for 77 individulal proteins collected from the Protein Data Bank, which is better than the previously reported accuracies.

DrugMiner: comparative analysis of machine learning algorithms for prediction of potential druggable proteins

“…Sequence properties determine whether a protein is targetable. Generally, the application of more relevant attributes is key to designing a simple and functional model that has better prediction performance [33,34]. Previous studies related to druggable protein predictions concentrated on only a few attributes, which are unlikely to represent all aspects of protein-drug interactions and also have issues related to limitations of estimations and assumptions [17,39].…”

“…It is well established that different attributes of proteins have a significant role in the interaction between proteins or between one or more drugs and a protein. These attributes vary from sequence-based features to physicochemical features [19,21,22,[24][25][26]33,34]. In this study, attributes were grouped into three categories: (i) Group 1: physicochemical properties of protein sequences, based on amino acid values in terms of their physicochemical parameters, such as length, weight, hydrophobicity, alpha helix, and so on; (ii) Group 2: amino acid composition, which was calculated based on the frequency of the 20 amino acid residues in the protein sequences; and (iii) Group 3: dipeptide composition, which was established based on the frequency of amino acid dimers in the protein sequences (Table 1).…”

Protein-protein Interfaces Integrated into Interaction Networks: Implications on Drug Design

“…Another parameter to characterize the interface is amino acid frequency [40,58,59]. The amino acid propensity of the interface is similar to interior of the protein [23,60].…”