Protein Subcellular Localization Prediction Based on SVM

Qinghua LIU,Yuping LAI,Hongwei DING,Zhijun YANG,Xiaolong Cui
DOI: https://doi.org/10.3778/j.issn.1002-8331.1802-0070
2019-01-01
Abstract:Based on feature fusion, combining amino acid composition, entropy density and autocorrelation coefficient to construct a 190 dimensional eigenvector for characteristic expression, this method can better express the protein structure information compared with the traditional method which only considers the amino acid composition information. It uses the Linear Discriminant Analysis(LDA)method to reduce the calculation complexity and increases the correlation between the samples. The support vector machine is selected as the classifier for positioning prediction. It uses the Jack-knife method to cross-check the gram-negative and gram-positive data sets. The experimental results show that the multi-feature combination method is superior to the traditional amino acid composition method and simple self-correlation coef-ficient method, and proves the validity of the new method.
What problem does this paper attempt to address?