Absorb metals cohort into v1 as v1.2 Path B extension

[realmarcin]

Summary

Consolidate the separate `metpo_communitymech_metals_v1` cohort (merged in PR #77) into v1 as a v1.2 Path B extension. Per the post-merge discussion, the metals cohort being separate was a judgment call rather than an enforced skill rule — the CHEBI-reuse policy difference can be documented as a section in v1's proposal.md without partitioning into a separate subset tag.

What changed

• Appended the 8 class rows and 3 property rows from `metpo_communitymech_metals_v1/` to v1's TSVs.
• Subset tag retagged from `metpo_communitymech_metals_2026_05` → `metpo_communitymech_2026_05` (single subset across the whole proposal).
• `METPO:1008000` parent corrected from `METPO:1007100` (microbial community) to `METPO:1000000` (peer top-level domain class). Same fix pattern as METPO:1007300 in PR Extend METPO v1: AtmosphereEnum + MediaRelationshipEnum (Path B) #76's Copilot review — community-relevant metal context is not a kind of microbial community, it's a peer concept.
• v1's `proposal.md` updated end-to-end:
  • Scope table — 3 new rows (MetalElementEnum CHEBI-reuse, RareEarthElementEnum CHEBI-reuse, MetalRelevanceEnum)
  • Top-level domain classes — now 5 rows (added METPO:1008000 for v1.2)
  • Hierarchy decisions — new subsection "MetalElementEnum and RareEarthElementEnum — CHEBI-reuse, no per-element leaves (v1.2)"
  • Predicate proposals — table now 19 rows with a Cohort column flagging v1 / v1.1 / v1.2
  • ID space — v1.2 block documented at 1008000-1008013 / 2008000-2008002 (deliberately at 1008000 rather than 1007320+ to visually separate the CHEBI-reuse classes from the all-leaves-minted v1/v1.1 classes)
  • Files — row counts updated (98 / 21 lines)
  • Change log — v1.2, 2026-05 entry covering the absorption and CHEBI-reuse policy
• `proposals/metpo_communitymech_metals_v1/` directory removed (its content lives in v1 now).

Final state on `main`

• One cohort: proposals/metpo_communitymech_v1/
• 96 class rows (was 88 in v1.1; +8 absorbed)
• 19 property rows (was 16 in v1.1; +3 absorbed)
• Same subset tag (metpo_communitymech_2026_05) on every row
• Coverage 67/67 enum permissible values across 12 enums
• Parent integrity 96/96 SC % references resolve

Test plan

• awk -F'\t' 'NF != 11' classes_robot.tsv returns nothing
• awk -F'\t' 'NF != 12' properties_robot.tsv returns nothing
• Per-enum coverage check: 67/67 across 12 enums
• Parent integrity check: 96/96
• (follow-up) robot template parses both TSVs cleanly

🤖 Generated with Claude Code

[realmarcin]

ROBOT validation — passes

Replicated the kg-microbe validate_with_robot pipeline (from scripts/extract_metpo_proposals.py:1643) against this branch's TSVs using local ROBOT 1.9.6:

robot template \
  --template proposals/metpo_communitymech_v1/metpo_proposal_classes_robot.tsv \
  --prefix "METPO: http://purl.obolibrary.org/obo/METPO_" \
  --prefix "biolink: https://w3id.org/biolink/vocab/" \
  --output /tmp/cm_classes.owl

robot template \
  --template proposals/metpo_communitymech_v1/metpo_proposal_properties_robot.tsv \
  --prefix "METPO: http://purl.obolibrary.org/obo/METPO_" \
  --prefix "biolink: https://w3id.org/biolink/vocab/" \
  --output /tmp/cm_props.owl

robot merge \
  --input /tmp/cm_classes.owl \
  --input /tmp/cm_props.owl \
  --output /tmp/cm_merged.owl

robot reason --reasoner ELK \
  --input /tmp/cm_merged.owl \
  --axiom-generators "SubClass EquivalentClass" \
  --output /tmp/cm_reasoned.owl

All four steps exit 0. The reasoned OWL contains:

• 97 METPO IRIs in classes.owl (96 minted classes + 1 external METPO:1000000)
• 36 METPO IRIs in props.owl (19 properties + 17 referenced domain/range classes)
• 117 METPO IRIs in reasoned.owl after ELK enrichment
• 0 owl:Nothing equivalences (no unsatisfiable classes)
• 0 inconsistency markers

robot report flags 4 ontology-metadata findings (3 ERROR missing title/license/description on the merged OWL itself, 1 WARN about the metpo_communitymech_2026_05 subset value not being formally declared as a NamedIndividual). All four belong to the merged-with-METPO output and disappear when the proposal lands in the real METPO ontology, which carries title/license/description and declares the subset; none reflect a problem with the proposed class or property content.

[Copilot]

Pull request overview

This PR consolidates the previously separate metals cohort (metpo_communitymech_metals_v1) into the main metpo_communitymech_v1 proposal as a v1.2 Path B extension, unifying everything under a single subset tag and removing the redundant proposal directory.

Changes:

• Appends the metals cohort’s 8 class rows and 3 property rows into v1’s TSV templates, using the unified subset tag (metpo_communitymech_2026_05).
• Updates v1 proposal.md to document the v1.2 extension (scope, hierarchy rationale for CHEBI reuse, predicates table, ID space, row counts, changelog).
• Removes the now-obsolete proposals/metpo_communitymech_metals_v1/ proposal files.

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
proposals/metpo_communitymech_v1/proposal.md	Documents the absorbed metals extension (v1.2) and updates scope/predicate/ID-space narrative.
proposals/metpo_communitymech_v1/metpo_proposal_properties_robot.tsv	Adds 3 new metals-related object properties (METPO:2008000–2008002) under the unified subset tag.
proposals/metpo_communitymech_v1/metpo_proposal_classes_robot.tsv	Adds 8 metals-related classes (METPO:1008000–1008013), including the corrected top-level parent and CHEBI-reuse grouping parents.
proposals/metpo_communitymech_metals_v1/proposal.md	Deleted (content absorbed into v1 proposal.md).
proposals/metpo_communitymech_metals_v1/metpo_proposal_properties_robot.tsv	Deleted (properties absorbed into v1 properties TSV).
proposals/metpo_communitymech_metals_v1/metpo_proposal_classes_robot.tsv	Deleted (classes absorbed into v1 classes TSV).

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.