PMC:4996382 / 25428-26981 JSONTXT

{"target":"https://pubannotation.org/docs/sourcedb/PMC/sourceid/4996382","sourcedb":"PMC","sourceid":"4996382","source_url":"https://www.ncbi.nlm.nih.gov/pmc/4996382","text":"5. Conclusions\nThe two virtual hybridization methods used in this study are useful to discriminate between organisms with highly similar genomes. The extended method utilizes the set of 13-nucleotide long probes but each probe is increased by four bases in the immediate vicinity of the two ends of the site recognized by the probe. Thus, 21-nucleotide (4 + 13 + 4) long comparison sequences were identified for each genome, which ensures that the formation of those heteroduplex was not random. \nSince the bacterial genome range in size between 0.5 and 9 million bp and 4.4 × 1012 different sequences can be formed by combining a sequence of 21 nucleotides, it is statistically highly unlikely for a sequence of this size to be found, by chance, in a bacterial genome and even less likely that it coincides in two bacteria. Therefore, these cases should correspond to sequences that have been conserved in these organisms. This suggests that, in a taxonomic tree, the separation of bacteria within the same genus is largely due to significant differences in the G + C content of the species involved. \nVirtual Hybridization software estimated a 0.000017 distance between the closest strains of Bacillus anthracis (this minimum value denotes the sensitivity to discriminate between highly similar strains, as the similarity between these species genomes is 99%). By comparing their virtual genomic fingerprints with the UFC we were able to detect these differences with the high potential and specific sites for each of these Bacillus anthracis strains.","divisions":[{"label":"Title","span":{"begin":0,"end":14}}],"tracks":[]}