Protein Fold Classification using Kohonen's Self-Organizing Map

Protein fold classification is an important problem in bioinformatics and a challenging task for machine-learning algorithms. In this paper we present a solution which classifies protein folds using Kohonen’s Self-Organizing Map (SOM) and a comparison between few approaches for protein fold classification. We use SOM, Fisher Linear Discriminant Analysis (FLD), K-Nearest Neighbour (KNN), Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) methods to classify three SCOP folds with six features (amino acid composition, predicted secondary structure, hydrophobicity, normalized van der Waals volume, polarity and polarizability). This paper has a novelty in the way of applying SOM to these six features, and also portrays the capabilities of SOM among the other methods in protein fold classification. The methods are tested on 120 proteins by applying 10-fold cross-validation technique and 93.33% classification performance is obtained with SOM.