Skip to content

H04 — TCM × Modern Rigor Methodology Lens (Stub)

Stub status. This card is committed at stub-level on 2026-05-05 to register the meta-hypothesis in the falsification-card directory and force the "what would kill this thesis" framing onto the TCM × modern rigor scope page. Full population (assumption stack, killshot menu, pre-committed thresholds, kill switches, failure-mode coverage map) is queued as Phase 2 P2-5 — see tcm-modern-rigor-intersection.md § Open Follow-Ups.

The pre-registration note on H01 (H01-ward-dual-cassette.md §Pre-registration) does not apply until this stub is upgraded to a full card. When the upgrade happens, the upgraded version is what gets pre-registered; the stub is informational scaffolding only.

Claim (provisional, stub-level)

The methodology lens defined in tcm-modern-rigor-intersection.md — six-rule discipline of (1) mechanism-mapped to chokepoints, (2) ChEMBL cross-checked, (3) bioavailability-honest with explicit gut-luminal embrace, (4) formula-decomposed for designed-coverage analysis, (5) standardized-extract-specified, (6) falsification-card-disciplined — produces actionable, gout-relevant findings beyond what either reductionist single-compound analysis (which loses the multi-component design) or holistic "alternative medicine" framings (which lose the rigor) produce in isolation.

The "actionable" bar is operationalized as one or more of:

  1. Mechanistically-grounded compound recommendations that survive ChEMBL cross-check at biologically achievable concentrations and map cleanly to a named chokepoint (NLRP3, ABCG2, URAT1, etc.).
  2. Multi-component formula decompositions that surface designed-coverage patterns (different components hitting different chokepoints) that justify the formula structure beyond folk tradition.
  3. Falsification-card-grade hypotheses for specific TCM-derived interventions (Smilax glabra as koji-payload candidate; Si Miao San as allopurinol adjunct; etc.) with pre-committed kill thresholds.
  4. Cross-domain insights — e.g., a TCM compound's classical "damp-heat" indication mapping onto a specific modern chokepoint (XO + ABCG2 modulation) suggests other "damp-heat" herbs in the same classical category may share the mechanism — testable hypothesis.

The thesis is gated on multilingual research access — Chinese-language clinical trial literature, ChiCTR registry data, Japanese Kampo literature, Korean traditional medicine sources. The 2026-05-05 conversation that triggered this page made explicit that Open Enzyme should ingest global multilingual sources by default rather than treating language as a "barrier" — the AI substrate (Claude, DeepSeek, Qwen) is multilingual at zero marginal cost. See CLAUDE.md for the global-multilingual-search default rule.

Assumption Stack (placeholder — to be populated in Phase 2 P2-5)

The full assumption stack will be populated after the Phase 2 lit scans (P2-1 classical formulas, P2-3 Smilax glabra deep-dive, P2-4 Si Miao San decomposition, P2-6 bioavailability characterization) and comp-011 (P2-2 ChEMBL cross-check) land. Anticipated load-bearing assumptions:

  1. The chokepoint map (nlrp3-exploit-map.md, gout-pathophysiology.md) is granular enough to capture TCM compound mechanisms (vs. forcing them into mismatched modern categories)
  2. ChEMBL's curated bioactivity dataset has sufficient coverage of TCM-relevant compounds (many natural products are sparsely curated in ChEMBL relative to medicinal-chemistry-derived compounds — this is documented in etc/chembl-cross-check.md)
  3. Modern Chinese clinical trial literature (ChiCTR registry) is sufficient quality to update wiki evidence tiers (Chinese RCTs have heterogeneous quality; this is empirically verifiable)
  4. The "embrace gut-luminal mechanisms" reframe (rule #3) actually predicts therapeutic effect for low-bioavailability TCM compounds (vs. the alternative explanation that they don't work at all). Comp-004's IC50 occupancy framework on curcumin / quercetin / EGCG is a partial validation; broader compound classes need similar treatment.
  5. The "designed coverage" interpretation of multi-component formulas (rule #4) is biologically defensible vs. being modern bias projection onto historical practice. Formula decomposition needs to falsify this too.
  6. Open Enzyme's bandwidth is sufficient to apply the methodology rigorously across enough compounds to surface non-obvious findings (vs. a cataloging exercise that adds compound pages without producing new platform-relevant insight)

Killshot Menu (placeholder — to be populated in Phase 2 P2-5)

Full killshot menu follows the H01 / H02 / H03 template. Anticipated highest-priority killshots:

  • Lit scan + ChEMBL cross-check first (Phase 2 P2-1 + P2-2). Cheapest move. If the systematic ChEMBL cross-check of the 8 candidate gout compounds reveals that most have NO curated bioactivity at biologically achievable concentrations against any chokepoint, the methodology lens produces nothing actionable — kill.
  • Si Miao San decomposition outcome (P2-4). If decomposition reveals all four herbs hit the same chokepoint (redundant rather than designed-coverage), the "formula decomposition surfaces designed coverage" claim weakens substantially.
  • Smilax glabra clinical evidence quality assessment (P2-3). If Chinese-language modern clinical literature on Smilax glabra is uniformly low-quality (small n, no controls, no hard endpoints), the "modern Chinese clinical trial literature is sufficient quality" assumption fails.
  • Cross-validation against existing wiki TCM-lineage compounds (oridonin, berberine, EGCG, theaflavins, curcumin, resveratrol). For each, retroactively apply the six-rule discipline. If the discipline produces no new insight beyond what the existing pages already capture, the methodology lens doesn't add value over the existing per-compound treatment.

Pre-Committed Thresholds (placeholder — to be populated in Phase 2 P2-5)

To be defined when the killshot menu is populated. Anticipated structure follows H01: declared Alive / Killed / Pending thresholds for each load-bearing claim.

Failure Modes Probed (placeholder — to be populated in Phase 2 P2-5)

To be populated. Anticipated relevant failure modes from linter-design.md §5: published-literature-gap, training-distribution bias (Western-research bias in the corpus), evidence-tier inflation (overstating "Animal Model" claims as "Supported" because TCM has long traditional use), the alternative-medicine-washing failure mode (modern-language dressing without underlying rigor — rule #6's discipline is the guard).

Partial closure on the ChEMBL-coverage-gap failure mode (added 2026-05-19). comp-013's original ChEMBL cross-check returned "5/9 compounds had no ChEMBL data" — the falsification card flagged this as a methodology-limit failure (ChEMBL coverage too sparse for TCM-relevant compounds to support rule #2). The 2026-05-19 traditional-name + traditional-formula re-scan (URAT1 rescan, XO rescan) partially closes this failure mode: mangiferin (CHEMBL3611008), apigenin (well-cataloged), acteoside (curated) are all in ChEMBL — the gap wasn't database coverage, it was seed-list construction at the herb level rather than the formula + marker level. The methodology fix is the 4-framing query matrix codified in manual-literature-mining.md §"Query-framing for non-Western compound classes". Rule #2 (ChEMBL cross-checked) survives intact; rule needing reinforcement is seed-list construction discipline upstream of rule #2.

Status

Stub. No killshot executed. No assumption stack pre-registered. Full hypothesis card is queued as Phase 2 P2-5 — see tcm-modern-rigor-intersection.md § Open Follow-Ups.

Survival count: 0.

Survival score: 0.0 (undefined until full card and first survived killshot).

Cross-References