MMseqs2 search results with Taxonomy ID using uniref30_2103 database and colabfold_envdb_202108 database

Dear friends, we got some trouble for MSA searching contain Taxonomy ID annotation.

Expected Behavior

We want to search using mmseqs2 with uniref30_2103 database and colabfold_envdb_202108 database. We want the results with the Taxonomy ID annotation for each sequence in MSA.

Current Behavior

We used a modified version of colabfold_search.sh, only with a modification from the convertail commend for Taxonomy ID as follows:

"${MMSEQS}" convertalis "${BASE}/prof_res" "${DBBASE}/${DB2}.idx" "${BASE}/res_pdb" "${BASE}/${DB2}.m8" --format-output query,target,taxid,taxname,taxlineage,fident,alnlen,mismatch,gapopen,qstart,qend,tstart,tend,evalue,bits,cigar --db-load-mode 2

After search, we found the "taxid,taxname,taxlineage" not in the MSA results. Is that because the pre-compiled dbs(uniref30_2103 database and colabfold_envdb_202108 database) does not contain the Taxonomy ID annotation? We also found the uniref30_2103_taxonomy.tar.gz updated on the website, please help us on how to make use of it, thank you~

Start with the modified colabfold_search.sh flow and its convertalis invocation, then inspect the uniref30_2103_taxonomy.tar.gz archive alongside the precompiled database contents. Determine whether the requested taxid, taxname, and taxlineage fields are available and how the archive is intended to be used; done means the MSA output includes those annotations or the limitation and required steps are documented.