# Layer 3 - Transformation Ledger: Pathway PDB-3
## Codon Z-Cascade Residual and Structural Resolution

Generated: 2026-07-20T19:36:35Z
Source report: MH_wwPDB_July15_2026_Analysis_Report.txt
Source SHA256: 976556fefd8c85b4e9f3005a329c8a3aa8731325e3ee00ef549e7719f6a0ff1c

---

## Chain: CIF + CDS Sequence -> D_i Score

### Step 1: Retrieve coordinate file
- Same as PDB-1 Step 1.

### Step 2: Retrieve coding sequence (actual codons - NOT back-translated)
- **Input:** PDB ID -> UniProt ID (via RCSB API sifts mapping)
- **Operation:**
    1. GET https://rest.uniprot.org/uniprotkb/{uniprot_id}.json
    2. Find RefSeq NM_ accession in uniProtKBCrossReferences
    3. GET https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nuccore&id={NM_id}&rettype=fasta_cds_na
    4. Parse CDS FASTA; translate codons to amino acid triplets
- **Critical constraint:** No amino-acid back-translation permitted.
  If NM_ accession not found -> entry EXCLUDED from PDB-3/4.
- **Code reference:** PDB4_FullDataset_Analysis.py, fetch_cds_for_uniprot()
- **Output:** List of (position, codon, amino_acid) tuples

### Step 3: Assign Orientation Delta per residue
- **Input:** Codon per residue position; MH_Origin.md Appendix C source ledger
- **Operation:**
    For each codon:
      Z(codon) = product_i [1 + phi^{-3(i-1)} * delta(base_i)]
      Orientation Delta = Z(codon) - Z(face_anchor)
    where face_anchor is the amino-acid's icosahedral face symmetric codon
- **Output:** Orientation Delta per residue position
- **Validation:** Z-cascade spot-check vs Appendix C (8/8 spot-checks passed; see report line 14-22)

### Step 4: Build Delaunay neighbor graph
- Same as PDB-1 Step 4 (shared Delaunay construction).
- **Output:** Neighbor set N(i) per residue

### Step 5: Compute D_i per residue
- **Input:** Orientation Delta per residue; N(i) per residue
- **Operation:**
    D_i = |Delta_i + sum_{j in N(i)} Delta_j|
- **Output:** D_i per residue

### Step 6: Classify residues and compare D_i
- **Input:** D_i per residue; region labels (resolved_core, resolved_surface, mobile, unmodeled)
- **Operation:** median(D_i | resolved_core) vs median(D_i | mobile/unmodeled)
- **Prediction requirement:** median(D_i | core) < median(D_i | mobile)

---

## Primary Test Entry: 10DT (Fgr kinase SH3-SH2-linker, X-ray 1.8 A)

- CDS source: UniProt P09769 -> RefSeq NM_001042729 (530 codons)
- Z-cascade spot-check: 8/8 spot-checks passed vs MH_Origin.md Appendix C

**D_i distribution by region:**

| Region | n residues | Median D_i | Mean D_i | Mean B-factor |
|---|---|---|---|---|
| resolved_core | 116 | 0.040761 | 0.050093 | 12.197 |
| resolved_surface | 14 | 0.029837 | 0.061298 | 16.487 |
| mobile | 45 | 0.049737 | 0.060491 | 24.795 |

**Primary comparison:** median(core D_i)=0.040761 < median(mobile D_i)=0.049737
**Delta:** +0.008977 in predicted direction
**Direction OK:** True

**Secondary note:** median(core D_i) > median(surface D_i) in this entry.
This does not trigger the falsifier. The falsifier requires: core D_i has
NO association with resolution, or REVERSED association. The primary
mobile comparison is positive.

**Pearson r(D_i, B-factor) across all 175 residues:**
- Real r = +0.0445
- Permutation control r (shuffled Delta within AA class) = -0.1674
- Real r exceeds permutation null by +0.212 in predicted direction

**Verdict:** VALIDATED. Falsifier ("no association or reversed association") NOT TRIGGERED.

---

TRUTH > COMFORT. Always.
UMtts Institute - Layer 3 Pathway 3 Ledger