Identification of DNA-binding proteins using support vector machines and evolutionary profiles

Abstract

Background: Identification of DNA-binding proteins is one of the major challenges in the field of genome annotation, as these proteins play a crucial role in gene-regulation. In this paper, we developed various SVM modules for predicting DNA-binding domains and proteins. All models were trained and tested on multiple datasets of non-redundant proteins.

SVM based prediction of RNA‐binding proteins using binding residues and evolutionary information

“…It was contrary to previously reported fact that dipeptide and four-part amino acid composition achieve better accuracy due to greater information content (Bhasin and Raghava, 2004a;Bhasin and Raghava, 2004b;Garg et al, 2005;Kumar et al, 2006). Similar trend in MCC was also reported with DNA-binding protein prediction method developed by our group (Kumar et al, 2007b).…”

“…In proteins, the number of amino acids (L) is not the same; hence we can not use PSI-BLAST PSSM directly to train SVM. In order to convert variable size L Â 21 dimension PSSM into fixed size 400 dimension input vector (PSSM-400) strategy used in DNA-binding proteins prediction method DNA binder was adopted (Kumar et al, 2007b). In short it was done in following steps: (a) all values of PSSM were normalized in range of 0-1 using formula (Value-minimum)/(maximum-minimum); (b) All rows belonging to the same amino acid were pooled together to form 20 matrices of size N AA Â 20, where N AA is the number of amino acid of type AA; (c) the summation of each column in new matrices (each daughter matrix) will produce a 20 dimensional vector.…”

Prediction of Wuchereria Bancrofti Troponin Antigenic Peptides: Application in Synthetic Vaccine Design to Counter Lymphatic Filariasis

“…MHC molecules have been well characterized in terms of their role in immune reactions. They bind to several peptide fragments generated after proteolytic cleavage of antigen [3]. This binding act like red flags for specific antigen and generates immune response against the parent antigen.…”

An Immunoinformatics Approach to Design Synthetic Peptide Vaccine from Dendroaspis polylepis polylepis Dendrotoxin-K(DTX-K)

“…MHC molecules have been well characterized in terms of their role in immune reactions. They bind to some of the peptide fragments generated after proteolytic cleavage of antigen [8]. This binding acts like red flags for antigen specific and to generate immune response against the parent antigen.…”