Splitting s__Bordetella pertussis?

Bacterial Taxonomy
1

Hi GTDB curation team,

I have been going through some basic data in GTDB r226 and found s__Bordetella pertussisto display uneven genome_size and GC distribution. Moreover, I found all isolates over 4.7Mb to have the NCBI_taxonomy of either Bordetella parapertussis or Bordetella bronchiseptica, while all isolates under 4.3Mb to have NCBI_taxonomy assigned to Bordetella pertussis. It appears like parapertussis and bronchiseptica share ANI of over 99%, and ANIs between them and pertussis are both over 98%.

image
image1107×685 45.7 KB

I am wondering if the large difference (~1.5Mb) in genome size justifies my suggestion of splitting s__Bordetella pertussis into perhaps 2 or 3 seperated GTDB species.
I am new to the GTDB community and may not be fully aware of the criterion for GTDB species curation process. Also, I am also not an expert in Bordetella. So I apologize if there are any mistakes.

Cheers,
David

2

Hi Dave,

Apologies for the slow response. GTDB uses ANI to delineate species. In general, GTDB considers strains that are >95% ANI to be from the same species. However, for strains that have already been recognized as different species we do relax this criterion to 97% ANI. You can find details on this in https://www.nature.com/articles/s41587-020-0501-8. Ultimately the assignment of strains to species is a matter of opinion. The “GTDB opinion” aims to follow a strict set of rules so that taxa at the same rank are generally comparable across the tree of life.

Personally, given the high similarity of these strains I would consider them to be different subspecies within a single species.

Cheers,
Donovan