Protein classification combining surface analysis and primary structure.

In this work, we propose a method for protein classification that combines the features extracted from both the primary structure and the surface analysis of a given protein. The surface analysis is used to find the amino acids that belong to the surface of the proteins. The most important finding of this work is to show that the features extracted from the amino acids that belong to the surface are useful in the classification process, since their contribution is partially independent from that of the features extracted from the whole primary structure; this property is used to build an ensemble of classifiers. The experimental results demonstrate the effectiveness of the proposed system. The idea is validated using three different data sets and three different feature extraction methods: 2-gram; residue couple; pseudo amino acid composition.