The evolutionary space model to be used for the metagenomic analysis of molecular and adaptive evolution in the bacterial communities

Abstract

Recent progress in metagenomics resulted in rapid accumulation of data on genetic diversity of microorganisms. This diversity is mostly represented by uncultured microorganisms, never described in regard to phenotype. Therefore, former phenotypic classification system of bacteria came out to be inapplicable to metagenomics and was thus replaced with a genotypic system, built upon the 16S rRNA gene. Metagenomics operates with nucleotide sequences instead of species. This shift in biodiversity assessment required a new classification system. In this study, we attempted to develop such a system. We call it the Evolutionary Space (ES). ES is a metric multidimensional space, where each point represents a single 16S rRNA sequence. These points are geometrically spaced according to the genetic distances between corresponding sequences. ES is aimed to represent genetic variability of all currently existing and theoretically predicted 16S rRNA genes and to identify obscure evolutionary patterns defining this enormous biodiversity. We used the mathematics of regular simplexes to identify dimensional properties of ES, constructed upon the distance matrix, derived from sequences stored in the SILVA database. It appeared to be 13D. After mapping the 16S rRNA database in ES a set of evolutionary patterns were observed on a series of ES slices (e.g., evolutionary hollows likely to represent probable positions of ancestral genes). Different bacterial phyla formed well-defined areas within the space. ES can also be used in the practice of metagenomic analysis (e.g., in the analysis of microbial communities’ succession). To show this we examined the dynamics of two microbiomes, exposed to salt stress in natural and artificial conditions. ES helps us to study these dissimilar microbiomes as mathematical objects possessing several geometric characteristics (e.g., shape, density, trajectory, and vector of community development). Although some essential biological questions could benefit from ES, its practical implementation requires collaboration in many fields of natural science, including bioinformatics, mathematics and, to some extent, astrophysics.