Understanding and Modeling Codon Usage Bias Of Metagenomic Samples

Microbes are everywhere in the world. However, understanding the attributes that characterize different microbial communities is still very challenging. Metagenomics is an emerging field that helps to understand the genetic and functional capabilities of microbial communities at the system level without any need for culturing. Metagenomics is also increasingly being recognized as a powerful tool to understand the community-level attributes that distinguish different communities. One such community-level attribute is codon usage which was widely used to characterize microbes at the species level and was shown to be related to various fundamental processes such as gene expression, metabolism, and horizontal gene transfer. However, little is known about the codon usage of microbial communities as a whole. Specifically, how the microbes from different ecological niches are related in terms of their codon usage is still not clear. Here in this study we considered protein-coding DNA sequences of more than 100 metagenomic samples collected from diverse environmental sources and analyzed their codon usage pattern. To compare and contrast codon usage of microbes in these communities here we mainly considered three matrices namely, codon adaptation index (CAI), effective number of codons (ENCs) and directional codon bias score (DCBS). Here we found cohesive signals in the codon usage of studied communities. Next, we calculated pair-wise sequence divergence between the microbes of different communities through average repetitive common substring index (ARSI) approach. Our results suggested that microbes from the similar environment tend to locate closer in the phylogenetic tree constructed based on ARSI score. When we correlated the codon usage pattern of chosen samples with their ARSI score we found consistent signals, however further studies are ongoing to understand the trend more clearly.