Extracting Features from Gene Ontology for the Identification of Protein Subcellular Location by Semantic Similarity Measurement

摘要

It is necessary to find a computational method for prediction of protein subcellular location (SCL). Many researches have focused on the topic. Among them, methods incorporated Gene Ontology (GO) achieved higher prediction accuracy. However the former method of extracting features from GO have some disadvantages. In this paper, to increase the accuracy of the prediction, we present a novel method to extract features from GO by semantic similarity measurement, which is hopeful to overcome the disadvantages of former method. Testing on a public available dataset shows satisfied results. And this method can also be used in similar scenarios in other bioinformatics researches or data mining process.