OBJECTIVE To investigate the species-level identification ability of different regions of 16S rDNA for common bacteria in pharmaceutical production environments.

METHODS The Shannon entropy was calculated for all type strain sequences recorded in the German Bacterial and Archaeal Nomenclature Database(List of Prokaryotic names with Standing in Nomenclature, LPSN) as well as those corresponding to commonly isolated bacteria in clean rooms, across different regions of 16S rDNA, to analyze the species-level identification ability of these regions. A similarity difference comparison was performed for all type strains recorded in LPSN, additionally, 403 representative bacterial isolates from pharmaceutical production environments were collected, sequenced for their 16S rDNA, and then analyzed based on similarity differences to explore the discrimination of identification results across different regions of 16S rDNA. Considering that different regions might have varying discrimination abilities for different genera and species, a correlation analysis was conducted for species with insignificant sequencing discrimination across the regions.

RESULTS Compared with other regions(533F–1100R region, 1100F–1492R region), the 27F–533R region of 16S rDNA exhibited the highest interspecies resolution. This region also showed the largest similarity difference between the first and second ranked sequence similarities. Expanding the alignment region did not significantly improve the discrimination between species.

CONCLUSION Prioritizing the use of the 27F–533R region of 16S rDNA for conducting species-level bacterial identification is an efficient and reliable approach.