Evaluation of genome similarities using independent components

Abstract

We propose the use of independent component analysis to find similarities of genomes. Considering different numbers of independent components, the complete linkage method was used to identify groups based on the estimated coefficients of the mixing matrix. The sequences analyzed correspond to the strains of the Mycobacterium tuberculosis genome, ten sequences were analyzed, obtained from the National Center for Biotechnology Information (NCBI, 2017). The GC-content of each sequence was evaluated using a sliding window of 10,000 bases. The clustering analysis using the independent components of the analyzed sequences was essential to verify the dissimilarity of the sequences.