Hybrid method for predicting protein location in cells Predicting Protein Subcellular Location Using Hybrid Approach
Shih-Hao Chen Han Wen Hsiao Xiaowen HongShih-Hao Chen Han-Wen Hsiao Wen-Hung Hsiao
Taichung Healthcare and Management University of biological information by
Institute of Bioinformatics, Taichung Healthcare and Management University
Biologically, the function of a protein is highly related to its subcellular localization.
It is necessary to develop a reliable method for protein subcellular localization prediction, especially when large-scale genome sequences are to be analyzed. Various methods have been proposed to perform the task . The results, however, are not satisfactory in terms of effectiveness and efficiency. A hybrid approach using naïve Bayesian classifier with information gain ratio and k-nearest neighbor classifier is presented. The total accuracy for a set of 17655 proteins can reach up to 91.5 %.
Key Words: subcellular localization, function prediction