The tool Get Data > UCSC Archaea would be the best way. The Table Browser at that site will accept a list of accessions and send fasta sequence directly to Galaxy. This will only work if the bacterial genome is included there and a track containing the protein sequences exists - so you will need to check.
If the data is not at the UCSC Archaea browser, there are two other choices I can think of:
- Go to ENA or NCBI directly, retrieve the sequences using their tools/methods, and upload the results to Galaxy after.
- Use Get Data > EBI SRA. Retrieval will be one sequence at a time. The fasta results can be concatenated within Galaxy after. This will be a tedious method and should probably only be considered as a last resort for your use case (which is not a common query to run on Galaxy).
Thanks! Jen, Galaxy team