关键词: 朴素贝叶斯分类器, 蛋白质相互作用界面, 序列谱, 残基溶剂可及表面积

Abstract: The identification of interface residues involved in protein-protein interactions （PPIs） has broad application in rational drug design and metabolic etc. A naïve Bayes classifier for PPIs prediction with features including protein sequence profile and residue accessible surface area is proposed. This method adequately uses the character of naïve Bayes classifier which assumes independence of the attributes given the class. The test results on a diversity dataset made up of only hetero-complex proteins achieve 68.1% overall accuracy with a correlation coefficient of 0.201， 40.2% specificity and 49.9% sensitivity in identify interface residues as estimated by leave-one-out cross-validation. This result indicates that the method performs substantially better than chance （zero correlation）. Examination of the predictions in the context of 3-dimensional structures of proteins demonstrates the effectiveness of this method in identifying protein-protein sites.

Key words: naï, ve Bayes classifier, protein-protein interactions, sequence profile, residue accessible surface area