DNA-MC: Tool for Mapping and Clustering DNA Sequences

Abstract

As a result of the increasing interest on genomic signal processing, there is the necessity of develop computational software that combines different tools for signal processing and automatic analysis. This paper presents a computational tool for mapping and clustering DNA sequences. Several DNA numerical representations, a feature extraction method, the K-means algorithm and different clustering evaluation metrics were implemented. This software allows to researchers to perform genomic signal analysis through a graphical user interface, without need programming skills. The tool is prepared to increase their capabilities by implementing different algorithms or modules. Also, a comparative analysis of eleven DNA numerical representation is presented. The results show that Electron-ion and Voss present the best performances when clustering genomic signals using K-means. The clusters quality was measure using the ARI metric.