A software tool for data mining of physicochemical properties of peptides

¹ Department of Computer Science and Mathematics, Faculty of Economics, Trakia University, 6000 Stara Zagora, Bulgaria
² Department of Mathematics, Physics and Information Technologies, Faculty of Economics, University of Food Technologies, 4002 Plovdiv, Bulgaria
³ Department of Computer Informatics, Faculty of Mathematics and Informatics, Plovdiv University “Paisii Hilendarski”, 4000 Plovdiv, Bulgaria
⁴ Department of Analytical Chemistry and Physical Chemistry, Technological Faculty, University of Food Technologies, 4002 Plovdiv, Bulgaria

^* Corresponding author: mterziyska@uft-plovdiv.bg

Abstract

Biologically active peptides (BAP) are increasingly in the focus of scientific research due to their widespread use in medicine, food and pharmaceutical industries. Researching and studying the properties of peptides is a laborious and expensive process. In recent years, in silico methods, including data mining or artificial intelligence, have been applied more and more to reveal biological, physicochemical and sensory properties of peptides. This significantly shortens the process of peptide sequences analysis. This article presents a software tool that uses a data mining approach to discover a number of physicochemical properties of a specific peptide. Working with it is extremely simple - it is only necessary to input the amino acid sequence of the peptide of interest. The software tool is designed to generate data in order to increase the classification and prediction accuracy, as well as to leverage the engineering of new amino acid sequences. This way, the proposed software greatly facilitates the work or scientific researchers. The software application is publicly available at www.pep-lab.info/dmpep.