Prediction of GPCRs with Pseudo Amino Acid Composition: Employing Composite Features and Grey Incidence Degree Based Classification
Abstract:G-protein coupled receptor (GPCR) is a protein family that is found only in the Eukaryotes. They are used for the interfacing of cell to the outside world and are involved in many physiological processes. Their role in drug development is evident. Hence, the prediction of GPCRs is very much demanding. Because of the unavailability of 3D structures of most of the GPCRs, the statistical and machine learning based prediction of GPCRs is much demanding. GPCRs are classified into family, sub family and sub-sub family levels in the proposed approach. We have extracted features using the hybrid combination of Pseudo amino acid, Fast Fourier Transform and Split amino acid techniques. The overall feature vector is then reduced using Principle component analysis. Mostly, GPCRs are composed of two or more sub units. The arrangement and number of sub units forming a GPCR are referred to as quaternary structure. The functions of GPCRs are closely related to their quaternary structure. The classification in the present research is performed using grey incidence degree (GID) measure, which can efficiently analyze the numerical relation between various components of GPCRs. The GID measure based classification has shown remarkable improvement in predicting GPCRs.
Keywords: BLAST & FASTA; Bayes network method; Chemokine; Frizzled & Smoothened receptors family; G-protein coupled receptor; GPCRs prediction; Hidden Markov models; Hybrid combination; Mathew's Selective Top Down method; Power spectral density; PseAAC; Rhodopsin; amino acids; fast fourier transforms; grey incidence degree measure; nearest neighbor classifier; principle component analysis; pseudo amino acid composition; quaternary structural classes; signal peptides; split amino acid
Document Type: Research Article
Publication date: September 1, 2011
- Protein & Peptide Letters publishes short papers in all important aspects of protein and peptide research, including structural studies, recombinant expression, function, synthesis, enzymology, immunology, molecular modeling, drug design etc. Manuscripts must have a significant element of novelty, timeliness and urgency that merit rapid publication. Reports of crystallisation, and preliminary structure determinations of biologically important proteins are acceptable. Purely theoretical papers are also acceptable provided they provide new insight into the principles of protein/peptide structure and function.