STRC h01 Phase 5e Mutant Re-Dock Delivery 2026-04-24

Pocket-stability gate for everything docked into Phase 5a WT snapshots. Re-docks 11 ligands (5 LEGACY_LEADS + 5 V2_HITS + tafamidis-analog) against 20 Phase 5d E1659A mutant MD snapshots. Box centre derived per-snapshot from Cα centroid of K1141 + pocket ring residues (1135, 1137, 1165, 1167, 1175); 18 Å cube; Vina exh=16. Settings identical to Phase 5b for direct comparability.

Launched 2026-04-23 ~19:36, delivered 2026-04-24 ~05:35 (~10 h wall, single-host CPU=8 parallel). Output: models/pharmacochaperone_phase5e_mutant_ensemble_redock.json.

Pocket-stability verdict — WT proxy validated ✅

6/11 ligands have a Phase 5b WT-ensemble comparator (5 LEGACY_LEADS + diflunisal positive control). The fenamic V2_HITS + tafamidis-analog have no Phase 5b WT-redock baseline (they post-date 5b).

Ligand	WT mean ΔG	MUT mean ΔG	ΔΔG	kd_ratio (mut/WT)	Verdict
naphthalene-2-COOH	−5.47	−5.44	+0.03	1.06×	no shift
indole-3-acetic-acid	−5.14	−5.04	+0.11	1.19×	no shift
nicotinic-acid	—	−3.95	—	1.29×	no shift
cyclopropane-phenyl-COOH	−5.02	−4.85	+0.17	1.33×	no shift
salicylic-acid	—	−4.27	—	1.96×	mutant weaker
diflunisal (positive control)	−5.43	−5.43	0.00	2.09×	mutant weaker

• Mean kd_ratio across 6 = 1.49×, max = 2.09× (diflunisal)
• All 6 fall under the +1 kcal/mol ΔΔG threshold from the v3b+5d gate.
• Diflunisal at the boundary (2.09× > 2× heuristic) — pocket geometry mostly preserved but slightly less accommodating to halogenated salicylate body. Within Vina’s pose-noise envelope (typical ±0.3–0.5 kcal/mol → ~2× Kd).

Conclusion: WT-based docking validated as proxy for E1659A. v3b’s 29 YELLOWs do not need wholesale re-docking on the mutant. Phase 5f deferred. WT ensemble is the operating receptor model going forward.

Chemistry verdict on mutant — RED across the board

Ligand	mean ΔG (kcal/mol)	std	best snap	Kd (µM)	f_PC@10µM
flufenamic-acid	−6.12	0.62	−7.20	32.7	0.117
niflumic-acid	−5.98	0.38	−6.83	41.1	0.098
tafamidis-analog	−5.85	0.50	−7.11	51.3	0.082
meclofenamic-acid	−5.49	0.46	−6.52	94.0	0.048
naphthalene-2-COOH	−5.44	0.37	−6.14	102.5	0.044
diflunisal (positive)	−5.43	0.43	−6.19	104.2	0.044
indole-3-acetic-acid	−5.04	0.26	−5.49	202.4	0.024
cyclopropane-phenyl-COOH	−4.85	0.42	−5.92	275.1	0.018
salicylic-acid	−4.27	0.30	−4.94	741.3	0.007
nicotinic-acid	−3.95	0.31	−4.63	1,276	0.004
sulfasalazine	−1.80	4.72	−6.70	47,925	0.000

• 0 GREEN / 0 YELLOW / 11 RED at MILD-MODERATE threshold f_PC ≥ 0.30
• Best mutant lead: flufenamic-acid Kd 32.7 µM (RED, fenamic-tox-blocked per STRC h01 Fenamic Scaffold Tox Audit 2026-04-23 — MD probe only)
• Tafamidis-analog mutant Kd 51 µM, f_PC 0.082 — bioisosteric pivot remains valid path but current single analog is not enough; needs the Phase 4h tafamidis-playbook library to materialize as ordered compounds
• Best snapshot pose at −7.11 (tafamidis-analog) and −7.20 (flufenamic) confirms upper-tail capacity exists; ensemble averaging dilutes due to per-snap pocket variance (std 0.4–0.6)
• Sulfasalazine: std 4.72 + best −6.70 → highly multimodal docking, likely too large for 18 Å box on subset of snapshots; mean unreliable, exclude from interpretation

Comparison with WT Phase 5b ceiling

Phase 5b best non-positive-control was naphthalene-2-COOH f_PC 0.047 on WT ensemble. Phase 5e best is flufenamic 0.117 on mutant ensemble — 2.5× improvement, but still 4.3× below the 0.30 MILD-MODERATE threshold and 6.4× below the v3b WT-ceiling f_PC 0.34 from the 12k library.

Translation: the new fenamic scaffold class beats the old shortlist on the mutant, but parent fenamates max out where v3b YELLOWs already are. The forward path is the same as before — Phase 3c v4 fragment-growing on the 3-amino-benzofuran-2-COOH scaffold.

Decision — next-step

Per the v3b+5d decision tree:

• (1) GREEN non-covalent → N/A (no GREENs)
• (2) GREEN covalent → N/A (no GREENs)
• (3) YELLOW → Phase 3c v4 fragment-grow on best YELLOW cluster (3-amino-benzofuran-2-COOH + polar tail subs + acid bioisosteres) — ACTIVATE
• (4) RED → Phase 3c v5 RFdiffusion de novo — defer until v4 RED

Phase 5f (v3b-29-on-mutant dedicated re-dock) stays deferred — 5e proved WT-ensemble adequate.

Phase 4h tafamidis-playbook library compounds (Tier 1 commercial: tafamidis #5, iododiflunisal 12, benzoxazole-3CN #1, benzoxazole-3OMe #2) become the v4 fragment-grow seed set.

Ranking delta

Tier A held. mech 3 / deliv 4 / misha_fit 4 all held.

• v3b YELLOWs validated against mutant pocket — no re-docking penalty needed; WT ensemble carries.
• Chemistry ceiling on parent fenamate scaffold confirmed RED; tafamidis-style bioisosteric optimization (Phase 4h playbook → Phase 3c v4 fragment-grow) is the operational forward path.
• Pharmacochaperone monotherapy-to-NORMAL still out of reach; adjunct-lever framing per Misha Compound-Het Therapy Stack Model intact.
• next_step: “Phase 5e mutant re-dock running (v3b YELLOW pocket-stability gate); then Phase 3c v4 fragment-grow on 3-amino-benzofuran-2-COOH scaffold” → “Phase 3c v4 fragment-grow on 3-amino-benzofuran-2-COOH scaffold (Phase 4h Tier-1 commercial seed set); Phase 5f deferred (5e validated WT ensemble as mutant proxy)”
• active_runs: cleared (Phase 5e delivered).
• lit_audit: fixed held; this is a compute delivery, no literature changes.

Connections

• [part-of] h01 hub
• [see-also] STRC h01 Phase 3c v3b + 5d Delivery 2026-04-24 — v3b chemistry ceiling + 5d MD source
• [see-also] STRC h01 Phase 5 MD Ensemble Rescoring 2026-04-23 — Phase 5b WT comparator
• [see-also] STRC h01 Phase 4h Tafamidis Playbook Library 2026-04-23 — v4 fragment-grow seed library
• [see-also] STRC h01 Fenamic Scaffold Tox Audit 2026-04-23 — fenamate parents are MD probes only
• [see-also] STRC h01 Parameter Provenance Audit 2026-04-25 — Vina/MM-PBSA gates documented as pipeline-specific
• Misha Compound-Het Therapy Stack Model
• STRC Hypothesis Ranking