what should our signature names look like for databases?

[ctb]

for genbank, we do --name-from-first, so we get output like this:

CP001941.1 Aciduliprofundum boonei T4...

for gtdb, we do trickier name setting, so we get output like this:

GCF_000025665 s__Aciduliprofundum boonei

with the main difference here being that the GCF_ identifier points to the identifier for the whole genome, not just the first sequence. That seems better.

We could add an optional identifier string to signatures. Hrm. Ref sourmash-bio/sourmash#268 for more such questions.

ref #7

[luizirber]

For wort I generated a name closer to GTBD:
GCF_000246355.1 Leptospira kirschneri serovar Mozdok str. 'B 81/7 type 3/Tsaratsovo' strain=B 81/7 type 3/Tsaratsovo, CLC_glsol0

I generate it from assembly_summary.txt, using assembly_accession, organism_name, infraspecific_name (if there is one) and finally a comma and asm_name.

This example is pretty much the worst case I found: long name, with ' in the middle (so I need to escape properly in the shell). But the crucial point is using GCF_000246355.1 in the first position, because --name-from-first in NCBI assemblies is a mess for our use cases.

[ctb]

we've standardized over the last two years on putting the identifier first, as above, and we use this for pretty everything (including sourmash taxonomy, and picklists). Everything seems to work fine ;). Closing as resolved 🎉 !