Subsequence profile map for protein fold recognition

Abstract

Information belonging to 3D structures of the proteins, which are the most fundamental macromolecules of life, plays a key role in bioinformatics studies. Protein fold recognition is considered as an important stage to determine 3D structures of the proteins. In this study, subsequence profile map (SPMap) is firstly used in protein fold recognition. The features extracted from each fold class are used in a two-layer approach to train classifiers for fold prediction. The classifier performance is evaluated with dataset using our proposed system, and 71.7% average accuracy rate is achieved.