Complete term structure for PAH perturbation in PKU model

[github-actions]

Summary

This PR adds the missing term structure and modifier to the PAH perturbation in the Harvey Whole-Body PKU Model computational model entry (Phenylketonuria.yaml). This makes the structure consistent with the other two computational models in the file.

Changes

• Added complete term structure for PAH perturbation:
  • term.id: HGNC:8582
  • term.label: PAH
  • modifier: ABSENT

Context

This change was cherry-picked from PR #57. The other changes in that PR contained fabricated evidence snippets that did not validate against PubMed abstracts:

• Narcolepsy: The new snippet for scheduled naps was not found in PMID:18830438 abstract
• Wilson's Disease: Both new snippets (dystonia phenotype and low copper diet) were not found in PMID:38731973 abstract

Only this structural improvement, which adds proper ontology term binding to match the pattern used in other models, has been preserved.

Validation

uv run linkml-validate -s src/dismech/schema/dismech.yaml -C Disease kb/disorders/Phenylketonuria.yaml
# No issues found

🤖 Generated by Dragon-AI Agent

Fixes #57

[claude]

Claude finished @cmungall's task in 2m 37s —— View job

---

PR Review: Complete term structure for PAH perturbation in PKU model

Todo List

• Read the modified Phenylketonuria.yaml file
• Identify critical issue: unresolved merge conflict
• Verify term structure consistency with other models
• Check biological validity of PAH perturbation in PKU
• Verify HGNC:8582 ontology term
• Analyze consistency issues
• Provide comprehensive review feedback

---

Critical Issues

1. Unresolved Merge Conflict 🚨

The file contains unresolved merge conflict markers on lines 249-257:

<<<<<<< dragon-ai/improve-evidence-snippets
    - preferred_term: PAH
      term:
        id: HGNC:8582
        label: PAH
      modifier: ABSENT
=======
  - preferred_term: PAH
>>>>>>> main

Impact:

• The file will NOT pass validation
• CI/CD checks will fail
• The PR cannot be merged until this is resolved

Resolution needed: The merge conflict must be resolved by removing the conflict markers and keeping the intended version.

---

2. Inconsistent Ontology Prefix Casing ⚠️

The PR uses uppercase HGNC:8582 (line 252), which is inconsistent with:

• The other two models in the same file (lines 223, 239): hgnc:8582 (lowercase)
• All other HGNC references across the entire knowledge base (200+ instances)

Evidence from codebase:

# Line 223 - Multi-compartment PKU FBA Model
id: hgnc:8582  # lowercase

# Line 239 - Recon3D with PAH knockout  
id: hgnc:8582  # lowercase

# Line 252 - Harvey Whole-Body PKU Model (THIS PR)
id: HGNC:8582  # UPPERCASE - inconsistent!

Recommendation: Change HGNC:8582 to hgnc:8582 to maintain consistency with the codebase standard.

---

Structural Review

Completeness Assessment

The PR correctly adds:

• ✅ term.id: HGNC:8582 (though case needs fixing)
• ✅ term.label: PAH
• ✅ modifier: ABSENT

This matches the structure of the first two computational models (lines 215-242), making all three models consistent in their perturbation specifications.

Indentation Analysis

The conflict shows an indentation change:

• Main branch: 2-space indent for the list item ( - preferred_term: PAH)
• This PR: 4-space indent with the item at 2-space ( - preferred_term: PAH)

The 4-space indent appears incorrect. YAML list items under perturbations: should be at 2-space indent relative to the parent key.

---

Biological and Clinical Validity ✅

PAH gene: Biologically correct for PKU

• PAH (phenylalanine hydroxylase) is the causative gene for phenylketonuria
• HGNC:8582 is the correct identifier for the PAH gene
• The modifier ABSENT is appropriate for modeling PAH deficiency

Evidence from file:

• Line 23: "Mutations in PAH gene cause deficient phenylalanine hydroxylase enzyme activity"
• Line 34: PMID:29025426 confirms PAH deficiency as core defect
• Line 49: PMID:34017006 describes "phenylalanine hydroxylase (PAH) deficiency"

---

Schema Conformance

Perturbation structure (from src/dismech/schema/dismech.yaml):

• ✅ Uses GeneDescriptor range (line 1487)
• ✅ modifier field is valid on Descriptor base class (line 846)
• ✅ ABSENT is a valid ModifierEnum value (line 173)
• ✅ Term binding to HGNC is appropriate for genes (lines 1905-1914)

The proposed structure is schema-compliant, pending resolution of the merge conflict and case correction.

---

Consistency Check

Comparing all three computational models in the file:

Model	Line	Term Structure	Modifier
Multi-compartment PKU FBA	221-225	✅ Complete	✅ ABSENT
Recon3D with PAH knockout	237-241	✅ Complete	✅ ABSENT
Harvey Whole-Body PKU (PR)	250-254	✅ Complete	✅ ABSENT

After this PR (once conflicts resolved), all three models will have consistent perturbation structures. ✅

---

Recommendations

Required Changes

1. Resolve merge conflict - Remove conflict markers and keep intended version
2. Fix ontology prefix casing - Change HGNC:8582 to hgnc:8582
3. Fix indentation - Ensure proper 2-space indent for list items under perturbations:

Suggested Actions

# Correct version should be:
perturbations:
  - preferred_term: PAH
    term:
      id: hgnc:8582  # lowercase to match codebase
      label: PAH
    modifier: ABSENT

Validation Commands

After fixes, run:

just validate kb/disorders/Phenylketonuria.yaml
just validate-terms-file kb/disorders/Phenylketonuria.yaml

---

Summary

Concept: ✅ Excellent - Adding complete term structure improves data quality and consistency

Implementation: ❌ Needs fixes - Merge conflict and casing inconsistency must be addressed

Biological validity: ✅ Correct - PAH with ABSENT modifier accurately represents PKU pathophysiology

Once the merge conflict and casing issues are resolved, this PR will be ready to merge.

[cmungall]

@claude fix botched merge conflict resolution

[claude]

Claude finished @cmungall's task —— View job

---

Todo List

• Read current state of kb/disorders/Phenylketonuria.yaml
• Identify and resolve merge conflict markers
• Fix casing consistency for HGNC terms
• Validate the changes
• Commit and push the fix

Merge Conflict Resolution

✅ Fixed the botched merge conflict in kb/disorders/Phenylketonuria.yaml

Changes Made

1. Resolved merge conflict markers on lines 249-257
2. Kept the complete term structure for the Harvey Whole-Body PKU Model:

   perturbations:
   - preferred_term: PAH
     term:
       id: hgnc:8582  # Fixed casing from HGNC:8582
       label: PAH
     modifier: ABSENT

3. Fixed casing consistency: Changed HGNC:8582 to hgnc:8582 to match the pattern used in the other two computational models (lines 223, 239)

Result

All three computational models in the file now have consistent perturbation structures:

• Multi-compartment PKU FBA Model ✅
• Recon3D with PAH knockout ✅
• Harvey Whole-Body PKU Model ✅ (now fixed)

The merge conflict has been resolved and changes pushed to the branch. The PR should now be ready to merge.

---

[cmungall]

@dragon-ai-agent please trigger QC workflow to validate this

[github-actions]

✅ QC Validation Complete

All QC checks passed successfully for the changes in this PR:

Schema Validation: ✓ Passed

• No structural issues found in Phenylketonuria.yaml

Term Validation: ✓ Passed

• All ontology term references (HGNC:8582 for PAH) are valid

Reference Validation: ✓ Passed

• No evidence snippets to validate (structural change only)

The added term structure for the PAH perturbation correctly follows the dismech schema pattern and matches the structure used in other computational models in the file. This PR is ready to merge.

---

🤖 QC validation performed by Dragon-AI Agent