Application of bioinformatics in the prediction and identification of potential antimicrobial peptides from Curcuma longa

In the current development stages, the antimicrobial peptides (AMPs) acquired from animals and plants had revealed as a novel ameliorative procedure as comparison to the conventional antimicrobial approach. The plants provides one of the richest natural sources of AMPs; they use AMPs to govern their own defense system against various biotic and abiotic stresses. These peptides have shown the potential antimicrobial properties against pathogenic microbes. In this study, in silico attempt have been made with the availability of plant proteome sequence - Curcuma longa - commonly known as turmeric which have shown promising feature of being antimicrobial in nature. Therefore, using stringent computational tools, C. longa proteome sequences were processed, analyzed and carried out to look for new peptides with likely potential antimicrobial activity. Firstly, major C. longa proteins were digested in silico by means of the three well known protease enzymes. Then, selection of digested peptide were carried out on the basis of results of different kinds of multidimensional statistics analysis such as support vector machines (SVM), random forest (RF), artificial neural network (ANN) and discriminant analysis (DA). Finally, predicted digested peptides were further characterized to examine different physicochemical properties and then compared with the patent antimicrobial peptides available at CAMP database. This study reveals a novel 28 potential peptides which may be considered a likely potential anti-microbial peptide activity.