This is how the file structure looks on my system:
$ tree --filelimit=10 data/databases/humann/
data/databases/humann/
├── chocophlan [30094 entries exceeds filelimit, not opening dir]
├── uniref
│ └── humann4_protein_database_filtered_v2019_06.dmnd
└── utility_mapping [21 entries exceeds filelimit, not opening dir]
$ ls data/databases/humann/chocophlan/ | head
SGB10000_pangenome90.fna.gz
SGB100018_pangenome90.fna.gz
SGB10001_pangenome90.fna.gz
SGB10002_pangenome90.fna.gz
SGB10003_pangenome90.fna.gz
SGB10004_pangenome90.fna.gz
SGB10005_pangenome90.fna.gz
SGB10006_pangenome90.fna.gz
SGB100078_pangenome90.fna.gz
SGB100087_pangenome90.fna.gz
$ ls data/databases/humann/utility_mapping/
map_eggnog_name.txt.gz map_ko_uniref90.txt.gz map_pfam_name.txt.gz metacyc_reactions_level4ec_only.uniref.bz2 vOct22_SGB_mapping.tsv
map_eggnog_uniref90.txt.gz map_level4ec_name.txt.gz map_pfam_uniref90.txt.gz mpa_vJan21_CHOCOPhlAnSGB_202103.tsv
map_go_name.txt.gz map_level4ec_uniclust90.txt.gz map_uniclust50_uniclust90.txt.gz mpa_vOct22_CHOCOPhlAnSGB_202403_species.txt
map_go_uniref90.txt.gz map_metacyc-pwy_name.txt.gz map_uniref90_name.txt.bz2 unipathway_pathways
map_ko_name.txt.gz map_metacyc-rxn_name.txt.gz metacyc_pathways_structured_filtered_v24_subreactions unipathway_uniprots.uniref.bz2
In my utility_mapping subdirectory, I see the mpa_vJan21_CHOCOPhlAnSGB_202103.tsv file instead of mpa_vOct22_CHOCOPhlAnSGB_202403.tsv, which I’d expected from the vOct22 CHOCOPhlAn SGB database. I’ve raised that potential issue in a separate thread with the maintainers, but there’s been no confirmation or fix so far.
Do you also have this mpa_vJan21_CHOCOPhlAnSGB_202103.tsv file in your utility_mapping subdirectory, or does your setup match the vOct22 naming instead?