Analysis of phylogenetic profiles using Bayesian decomposition

Antibiotic resistance together with the side effects of broad spectrum antibacterials make development of targeted antibiotics of great interest. To meet the problem of identifying potential targets specific to some genuses, a dataset comprising a series of phylogenetic profiles was built for a series of pathogenic bacteria of interest. The profiles are the highest BLAST scores for genes compared to selected genes of E. coli and M. tuberculosis. The dataset reflects the past evolution of those genes due to adaptation to specific niches, marked by lateral gene transfer, duplication and mutation of existing genes, or merging of existing genes. Genes that function together will be constrained to evolve together, to maintain viability in the organism. However, a given gene may have a role in multiple functional groups through the evolutionary process. Analysis using Bayesian decomposition helps to retrieve those relationships by retrieving fundamental patterns related to the evolutionary retained functions.