PLoc-Euk: An Ensemble Classifier for Prediction of Eukaryotic Protein Sub-cellular Localization

摘要

Protein Sub-Cellular Localization is very important information as they play a crucial role in their functions. Thus, prediction of protein Sub-Cellular Localization has become very promising and challenging problem in the field of Bioinformatics. Recently, a number of computational methods based on amino acid compositions or on the functional domain or sorting signal. But, they lack of contextual information of the protein sequence. In this paper, an ensemble classifier, PLoc-Euk is proposed to predict sub-cellular location for the eukaryotic proteins which uses multiple physico-chemical properties of amino acid along with their composition. PLoC-Euk aims to predict protein Sub-Cellular Localization in eukaryotes across five different locations, namely, Cell Wall, Cytoplasm, Extracellular, Mitochondrion, and Nucleus. The classifier is applied to the dataset extracted from http://www.bioinfo.tsinghua.edu.cn/~guotao/data/and achieves 73. 37% overall accuracy.