Title :
Prediction of Protein Subcellular Localization Using EDA Based Ensemble Classifiers
Author_Institution :
Sch. of Inf. Technol., Shandong Women´´s Univ., Jinan, China
Abstract :
The function of protein is closely correlated with its sub cellular locations. New composed proteins can perform normal biological function only after they are translocated to correct sub cellular locations. In this paper, a new selective ensemble classifiers based on EDA algorithm has been proposed. In the method, pseudo amino acid composition was firstly applied to form the protein feature sets, then 10 neural networks is generated to learn the subsets which are re-sampling from feature subsets with PSO algorithm. At last, appropriate classifiers are selected to construct the prediction committee with EDA algorithm. Experiment shows that the proposed method produces the best prediction accuracy than the other methods on SNL6 database.
Keywords :
bioinformatics; neural nets; particle swarm optimisation; pattern classification; proteins; EDA based ensemble classifiers; PSO algorithm; bioinformatics; estimation of distribution algorithm; feature subsets; neural networks; normal biological function; protein subcellular localization; pseudo amino acid composition; selective ensemble classifiers; sub cellular locations; Accuracy; Amino acids; Classification algorithms; Neural networks; Prediction algorithms; Protein sequence;
Conference_Titel :
Computer Science & Service System (CSSS), 2012 International Conference on
Conference_Location :
Nanjing
Print_ISBN :
978-1-4673-0721-5
DOI :
10.1109/CSSS.2012.460