Skip to content

Relax taxon constraint on UBERON:0001002 (cuticle)#3695

Open
ai4c-agent[bot] wants to merge 1 commit intomasterfrom
dragon-ai-agent-issue-3694-run811
Open

Relax taxon constraint on UBERON:0001002 (cuticle)#3695
ai4c-agent[bot] wants to merge 1 commit intomasterfrom
dragon-ai-agent-issue-3694-run811

Conversation

@ai4c-agent
Copy link
Copy Markdown

@ai4c-agent ai4c-agent Bot commented Apr 25, 2026

Summary

Removes the in_taxon NCBITaxon:6072 (Eumetazoa) constraint from the generic cuticle grouping class (UBERON:0001002), per @cmungall's main recommendation in #3694.

The class is already flagged as a non-homologous grouping (subset: grouping_class, subset: non_informative) and its own definition acknowledges this: "Cuticles are non-homologous, differing in their origin, structure and chemical composition". The Eumetazoa constraint was therefore an editorial choice rather than a biological assertion, and it caused downstream taxon constraint violations: GO terms like cuticle development reference UBERON:0001002 in their logical definitions, propagating the constraint to plant (Arabidopsis) annotations.

This is the least-disruptive fix as requested. It does render the generic cuticle class quite abstract — clade-specific subclasses (chitin-based cuticle, nematode cuticle, etc.) should be preferred where possible.

Changes

  • Removed relationship: in_taxon NCBITaxon:6072 ! Eumetazoa
  • Updated comment to record the rationale and point users to clade-specific subclasses
  • Added term_tracker_item linking to incorrect taxon constraints #3694
  • (subset: grouping_class was already present)

Resolves

Follow-ups (separate PRs)

Per @cmungall's request, separate PRs will follow improving the invertebrate cuticle representation, e.g.:

  • Add annelid cuticle (Lophotrochozoa, collagen-based, non-molted)
  • Broaden chitin-based cuticle from Arthropoda to Panarthropoda (or add ecdysozoan cuticle parent)
  • Add tardigrade / onychophoran / scalidophoran / nematomorph cuticle subclasses

Test plan

  • Verify ROBOT/ODK build passes (taxon constraint check, reasoning, etc.)
  • Confirm UBERON:0001002 no longer asserts in_taxon Eumetazoa
  • Confirm clade-specific subclasses (UBERON:0001001, UBERON:0003049) still carry their own taxon constraints

🤖 Generated by @dragon-ai-agent

Removes the `in_taxon NCBITaxon:6072` (Eumetazoa) constraint from the
generic 'cuticle' grouping class. The class is by construction
non-homologous (the text definition itself states 'Cuticles are
non-homologous, differing in their origin, structure and chemical
composition'), so a hard taxon constraint was editorially debatable to
begin with. More importantly, the constraint caused downstream taxon
constraint violations: GO terms such as 'cuticle development' have
logical definitions referencing UBERON:0001002, which propagated the
Eumetazoa restriction to plant (Arabidopsis) annotations.

This is the least-disruptive fix per @cmungall in
#3694, and resolves the violations reported in
geneontology/go-ontology#30555 without requiring GO-side changes.

The `grouping_class` subset is already present. Updated the comment to
explain the rationale and to point users at clade-specific subclasses
(e.g. chitin-based cuticle, nematode cuticle), and added a
term_tracker_item link to the issue.

Improvements to the invertebrate cuticle representation
(annelid cuticle, broadening chitin-based cuticle to Panarthropoda,
etc.) will be handled in separate PRs as requested.

Signed-off-by: @dragon-ai-agent
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

incorrect taxon constraints

0 participants