Abstract

Info

Author(s):: XU Shutan1; 2; WANG Junhao1; 2; CHEN Ming1; 2; 1. College of Information Technology, Shanghai Ocean University, Shanghai 201306, China 2. Key Laboratory of Fisheries Information, Ministry of Agriculture and Rural Affairs of the Peoples Republic of China, Shanghai 201306, China

Keywords:: Keywords: protein-GDP binding site position-specific scoring matrix under-sampling sliding window support vector machine

Abstract:: Abstract: The prediction of protein-GDP (Guanosine Diphosphate) binding site is significant for protein function annotation and new drug discovery. A sequence-based protein-GDP binding site prediction method is proposed for improving the accuracy of protein-GDP binding site prediction. The method uses a position-specific iterative algorithm for multiple sequence comparison to obtain a position-specific scoring matrix, selects the feature vector of each residue in the protein sequence through the mirror residue-based variable sliding window, solves the imbalance problem of the positive and negative samples of the data set using CNMW (Clustering NearMiss-2 Weighted) under-sampling, and finally realizes the prediction via support vector machine. The experimental results showed that compared with traditional methods, the proposed method has a significantly higher Matthews correlation coefficient, indicating its effectiveness and feasibility.