Protein structural classification using orthogonal transformation and class-association rules.

Protein structure classification and comparison is a central area in the field of bioinformatics. Rapidly increasing protein structure databases commonly suffer from the 'curse of dimensionality', necessitating the development of the dimensionality reduction of structural information prior to its classification. We propose a novel automated algorithmic framework for three-dimensional structure-based classification of proteins using orthogonal transformation of the geometric shape descriptors derived from protein structures, and then employing an association rule-based supervised clustering approach. The proposed computational framework demonstrates, on two different data sets, the applicability of association rule discovery-based classification of structural descriptors for protein fold classification with improved sensitivity.