K-mer-Based Motif Analysis in Insect Species across Anopheles, Drosophila, and Glossina Genera and Its Application to Species Classification
Year: 2019
Type: article
Abstract: Short k-mer sequences from DNA are both conserved and diverged across species owing to their functional significance in speciation, which enables their use in many species classification algorithms. In the present study, we developed a methodology to analyze the DNA k-mers of whole genome, 5′ UTR, intron, and 3′ UTR regions from 58 insect species belonging to three genera of Diptera that include Anopheles, Drosophila, and Glossina. We developed an improved algorithm to predict and score k-mers based on a scheme that normalizes k-mer scores in different genomic subregions. This algorithm takes advantage of the information content of the whole genome as opposed to other algorithms or studies that analyze only a small group of genes. Our algorithm uses k-mers of lengths 7–9 bp for the whole genome, 5′ and 3′ UTR regions as well as the intronic regions. Taxonomical relationships based on the whole-genome k-mer signatures showed that species of the three genera clustered together quite visibly. We also improved the scoring and filtering of these k-mers for accurate species identification. The whole-genome k-mer content correlation algorithm showed that species within a single genus correlated tightly with each other as compared to other genera. The genomes of two Aedes and one Culex species were also analyzed to demonstrate how newly sequenced species can be classified using the algorithm. Furthermore, working with several dozen species has enabled us to assign a whole-genome k-mer signature for each of the 58 Dipteran species by making all-to-all pairwise comparison of the k-mer content. These signatures were used to compare the similarity between species and to identify clusters of species displaying similar signatures.
DOI: 10.1155/2019/4259479
Authors: Matyas Cserhati, Peng Xiao, Chittibabu Guda
Institution: University of Nebraska Medical Center
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: hybrid
Estimated APC paid: $2,550
OpenAlex ID: W2986205472