When curating unaligned external prefixes, many are either out of scope or not possible to find minimum metadata and therefore should be explicitly excluded from the Bioregistry.

**Warning ** This is a first draft of a tutorial on explicitly curating non-alignments. It’s subject to change and additional polishing.

  1. Identify a prefix of interest from BioPortal (or another) curation sheet (e.g., here)
  2. Decide if it’s out of scope:
    • Not enough information inside external registry’s record to follow-up on what it is (i.e., 000467 in CHEMINF)
    • Is there evidence of this resource being used somewhere? Don’t add resources that are one-off (e.g., many ontologies in BioPortal) as this pollutes the prefix pool and makes the burden of finding things harder
    • Double mapping to a Bioregistry prefix (there isn’t currently a way to curate these)
    • Not about biomedical, chemical, clinical, or life sciences or meta-sciences
  3. If out of scope, add to the SKIP list, either in the corresponding alignment python module or associated processing configuration (TODO this needs to be better standardized)

The PR #41 demonstrates curating CHEMINF mappings including adding three explicit blacklists.