A Review on Protein Structure Classification along with a Proposed Classifier Using Data Mining Techniques

Protein structure classification is an approach to classify unknown protein into its class using the structural properties of the protein. As new protein structures are rapidly increasing, the need for efficient and automated data mining techniques for classifying proteins into classes with high accuracy and low computational time is also increasing. There are numerous techniques which are available for Protein structural classification, some of them have been reviewed in this paper. To execute the same, various researchers have proposed many efficient data mining approaches like representing protein structures in a hierarchy tree structure, random forest approach, amino acid residue’s averaged chemical shifts (ACS), Hidden Markov Model (HMM), etc. A comparative analysis is also performed with the limitations of above approaches to find the gap of research in this field. Finally, a new classification model is proposed based on the limitation of the above approaches, which can classify the protein with better accuracy and low computational time.