Schema: per-taxon abundances + reusable CommonTaxon gene records

CLAUDE.md

@@ -35,7 +35,9 @@ src/communitymech/
 │   └── reference_validator.py   # Validates evidence items in YAML files
 └── cli.py                       # Entry point (not yet implemented)
 
-kb/communities/                  # 60 curated community YAML files
+kb/communities/                  # curated community YAML files (root class MicrobialCommunity)
+kb/taxa/                          # reusable per-taxon gene records (root class CommonTaxon);
+                                  #   referenced from taxonomy[].common_taxon; `just validate-taxa`
 conf/oak_config.yaml             # OAK ontology adapter config (NCBITaxon, ENVO, CHEBI, GO)
 references_cache/                # Cached PubMed abstracts (committed for reproducibility)
 scripts/                         # Utility scripts for curation (not part of package)

conf/id_label_targets.yaml

@@ -92,6 +92,18 @@ targets:
       - { id: "NCBITaxon:1807132", label: "Candidatus Phormidium alkaliphilum", reason: "absent from current OAK ncbitaxon snapshot and kg-microbe ncbitaxon TSV" }
       - { id: "NCBITaxon:3050471", label: "Stenotrophomonas goyi", reason: "absent from current OAK ncbitaxon snapshot and kg-microbe ncbitaxon TSV" }
 
+  # data inputs (reusable per-taxon gene records) — taxon_term.term (NCBITaxon)
+  # and gene go_terms (GO) must be canonical. genome {id,label} are NCBI Assembly
+  # accessions (no CURIE prefix → benign SKIPPED_NO_ADAPTER) and the record's own
+  # CommunityMech:taxon:* id is an ignored prefix, so only the ontology terms are
+  # checked here.
+  - name: taxa_yaml
+    kind: yaml
+    glob: "kb/taxa/*.yaml"
+    policy: canonical
+    pairs:
+      - [id, label]
+
   # data product: KGX node export carries (id, name) — canonical OR synonym
   - name: kgx_nodes
     kind: tabular

justfile

@@ -23,6 +23,20 @@ validate-all:
         uv run linkml-validate -s src/communitymech/schema/communitymech.yaml "$file"
     done
 
+# Validate the reusable per-taxon gene records (kb/taxa/) against CommonTaxon.
+# These files have CommonTaxon as their root, not MicrobialCommunity, so the
+# target class must be given explicitly.
+validate-taxa:
+    #!/usr/bin/env bash
+    set -uo pipefail
+    rc=0
+    for file in kb/taxa/*.yaml; do
+        echo "Validating $file..."
+        uv run linkml-validate -s src/communitymech/schema/communitymech.yaml \
+            --target-class CommonTaxon "$file" || rc=1
+    done
+    exit $rc
+
 # Strict in-process validation in *closed* mode (rejects unknown fields).
 # Emits reports/instance_validation_failures.tsv and exits 1 on any ERROR.
 # Catches the same drift class that gave CultureMech 59k silent errors;

kb/communities/Shewanella_Geobacter_Exoelectrogenic_Biofilm_Community.yaml

@@ -51,6 +51,7 @@ taxonomy:
       label: Shewanella oneidensis
     notes: Exoelectrogenic proteobacterium in the defined anode biofilm community.
   abundance_level: ABUNDANT
+  common_taxon: CommunityMech:taxon:000001
   functional_role:
   - SYNTROPHIC_PARTNER
   - CROSS_FEEDER
@@ -73,6 +74,7 @@ taxonomy:
     notes: Exoelectrogenic Geobacter member that was detected in the planktonic phase of mixed-culture
       reactors as well as in the anode-associated community.
   abundance_level: ABUNDANT
+  common_taxon: CommunityMech:taxon:000002
   functional_role:
   - SYNTROPHIC_PARTNER
   - CROSS_FEEDER

kb/taxa/Geobacter_sulfurreducens.yaml

@@ -0,0 +1,57 @@
+id: CommunityMech:taxon:000002
+taxon_term:
+  preferred_term: Geobacter sulfurreducens
+  term:
+    id: NCBITaxon:35554
+    label: Geobacter sulfurreducens
+genomes:
+- id: GCF_000007985.2
+  label: ASM798v2
+  notes: RefSeq reference assembly for Geobacter sulfurreducens PCA.
+genes:
+- gene_id: KEGG:gsu:GSU1496
+  gene_symbol: pilA
+  locus_tag: GSU1496
+  product: Type IV pilin; structural subunit of conductive (electrically conductive) pili / e-pili
+  genome: GCF_000007985.2
+  go_terms:
+  - id: GO:0009055
+    label: electron transfer activity
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Conductive pili (e-pili) mediate long-range extracellular and direct
+    interspecies electron transfer (DIET) to partner methanogens and to
+    Fe(III) oxides/electrodes.
+- gene_id: KEGG:gsu:GSU2504
+  gene_symbol: omcS
+  locus_tag: GSU2504
+  product: Outer-surface hexaheme c-type cytochrome associated with conductive pili
+  genome: GCF_000007985.2
+  go_terms:
+  - id: GO:0009055
+    label: electron transfer activity
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Cytochrome filament/decoration required for extracellular electron transfer
+    to Fe(III) oxides and for direct interspecies electron transfer to partners.
+- gene_id: KEGG:gsu:GSU2076
+  gene_symbol: omcZ
+  locus_tag: GSU2076
+  product: Outer-surface octaheme c-type cytochrome
+  genome: GCF_000007985.2
+  go_terms:
+  - id: GO:0009055
+    label: electron transfer activity
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Cytochrome essential for high-density current production at electrodes
+    (extracellular electron transfer to anodes).
+notes: >
+  Reusable taxon record capturing the conductive-pili and multiheme-cytochrome
+  genes that underpin G. sulfurreducens' electron-transfer / syntrophic-partner
+  role in DIET cocultures, Fe(III)-reduction, and bioanode communities. Gene/role
+  claims are literature-standard; per-interaction EvidenceItems can be added
+  following the community-record evidence protocol.

kb/taxa/Shewanella_oneidensis_MR1.yaml

@@ -0,0 +1,67 @@
+id: CommunityMech:taxon:000001
+taxon_term:
+  preferred_term: Shewanella oneidensis MR-1
+  term:
+    id: NCBITaxon:211586
+    label: Shewanella oneidensis MR-1
+genomes:
+- id: GCF_000146165.2
+  label: ASM14616v2
+  notes: RefSeq reference assembly for Shewanella oneidensis MR-1.
+genes:
+- gene_id: KEGG:son:SO_1778
+  gene_symbol: mtrC
+  locus_tag: SO_1778
+  product: Outer-membrane decaheme c-type cytochrome (terminal extracellular electron-transfer reductase)
+  genome: GCF_000146165.2
+  go_terms:
+  - id: GO:0009055
+    label: electron transfer activity
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Terminal reductase of the Mtr extracellular electron transfer pathway; transfers
+    electrons to extracellular acceptors (e.g. Fe(III)/Mn(IV) oxides, electrodes),
+    underpinning direct interspecies / electrode electron transfer interactions.
+- gene_id: KEGG:son:SO_1776
+  gene_symbol: mtrB
+  locus_tag: SO_1776
+  product: Outer-membrane beta-barrel that embeds the MtrA/MtrC cytochrome module
+  genome: GCF_000146165.2
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Forms the outer-membrane conduit of the MtrCAB porin-cytochrome complex enabling
+    electron egress to extracellular acceptors.
+- gene_id: KEGG:son:SO_1777
+  gene_symbol: mtrA
+  locus_tag: SO_1777
+  product: Periplasmic decaheme c-type cytochrome of the Mtr pathway
+  genome: GCF_000146165.2
+  go_terms:
+  - id: GO:0009055
+    label: electron transfer activity
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Conducts electrons across the outer membrane within the MtrCAB complex during
+    extracellular electron transfer.
+- gene_id: KEGG:son:SO_4591
+  gene_symbol: cymA
+  locus_tag: SO_4591
+  product: Inner-membrane tetraheme c-type cytochrome (menaquinol oxidase) feeding the Mtr pathway
+  genome: GCF_000146165.2
+  go_terms:
+  - id: GO:0009055
+    label: electron transfer activity
+  supports_roles:
+  - SYNTROPHIC_PARTNER
+  supports_interaction: >
+    Branch point that delivers electrons from the menaquinone pool to periplasmic
+    and outer-membrane cytochromes for extracellular electron transfer.
+notes: >
+  Reusable taxon record capturing the Mtr extracellular-electron-transfer pathway
+  genes that underpin S. oneidensis MR-1's electron-donor role in microbial fuel
+  cell, metal-reduction, and DIET-style communities. Gene/role claims are
+  literature-standard; per-interaction EvidenceItems can be added following the
+  community-record evidence protocol.