Application of novel alignment-free sequence descriptors in Zika virus characterization

The Zika virus is spread by the Aedes aegypti mosquito which is prevalent in tropical climate regions, but is spreading northwards following global warming trends. To understand the nature of the virus we undertook a special bioinformatics study of its complete RNA sequence from African and non-African, i.e., Asian, Pacific and American, sources. We know that viral sequences mutate much more rapidly than, say, mammalian sequences. We found that the African Zika viral sequences were distinctly separate compared to the non-African sequences.

This difference is evident dramatically from two graphs of the Zika genome, one from Central African Republic and another from Brazil. The graphs are representations of the Zika RNA sequences through our model where the nucleotides of a sequence are plotted sequentially on a 2-dimensional grid by a set algorithm. Such a representation allows us to project a visual rendering of the distribution of nucleotides in a sequence and also to quantitatively index a sequence through numerical sequence descriptors and measure similarities and dissimilarities among sequences. While the normal practice is to use various alignment techniques like BLAST, CLUSTALW, etc. to measure such differences, our approach allows for a novel method of genome characterization called alignment-free measures and is therefore independent of several assumptions. Results such as ours provide more focus to surveillance of the spread and mutations of the Zika virus and hopefully can contribute to the eventual development of suitable drugs and vaccines that target the expression of specific Zika virus genes or the products expressed by these genes.