Spaces instead of underscores in FASTA download

[peterjc]

http://oomycetedb.cgrb.oregonstate.edu/search.html says:

There are multiple releases of the database. Usually, you should use the latest release, unless you are trying to reproduce a previous analysis. You can download the entire database here or just a subset.

The FASTA header format

The database is a FASTA file with headers in the following format:

name=Aphanomyces_invadans|strain=NJM9701|ncbi_acc=KX405005|ncbi_taxid=157072|oodb_id=13|taxonomy=cellular_organisms;Eukaryota;Stramenopiles;Oomycetes;Saprolegniales;Saprolegniaceae;Aphanomyces_invadans

The following fields are present:

“name”: The binomial species name of the organism with spaces replaced by underscores.
“strain”: The name of the strain/isolate if available. If the strain is not available, it is left empty.
“ncbi_acc: The NCBI accession number for the sequence submitted to genbank. Note that the version number (the number at the end, after the period) is not included.
“ncbi_taxid”: The NCBI taxonomy id. This can be looked up using the NCBI accession number.
“oodb_id”: This is the unique numeric ID specific to OomyceteDB.
“taxonomy”: The taxonomic classification separated by semicolons. This classification is curated by us and is not the taxonomic classification from NCBI associated with the NCBI taxid.

However, the downloads I made in the last few days still have spaces in the species name (as shown in both the name= and taxonomy= fields, counter to the example). e.g.

$ grep NJM9701 oomycetedb_whole_2022_06_28_08_47_34.fa 
>name=Aphanomyces invadans|strain=NJM9701|ncbi_acc=KX405005|ncbi_taxid=157072|oodb_id=13|taxonomy=cellular_organisms;Eukaryota;Stramenopiles;Oomycetes;Saprolegniales;Saprolegniaceae;Aphanomyces invadans

$ grep NJM9701 oomycetedb_whole_2022_07_06_03_05_06.fa 
>name=Aphanomyces invadans|strain=NJM9701|ncbi_acc=KX405005|ncbi_taxid=157072|oodb_id=13|taxonomy=cellular_organisms;Eukaryota;Stramenopiles;Oomycetes;Saprolegniales;Saprolegniaceae;Aphanomyces invadans

This is important as most FASTA parsers will take the first word as the identifier, i.e. breaking at the first space.