Identifying protein submitochondrial location by using features of sequence

The function of proteins is closely related to it´s subcellular localization. With the number of sequences entering into databanks rapidly increasing, it is highly desirable to predict a protein subcellular localization from its amino acid sequence. In this paper, the amino acid composition, N-terminal region of protein sequence, the amino acid hydropathy composition and gene ontology are selected as feature parameters by using support vector machine. The predictive results show that our method is efficient to predict the protein submitochondrial localization.