Cardio21
PYRIN-PLAQUE Digital Twin investigates whether MEFV/pyrin inflammasome biology can be separated from the saturated NLRP3 atherosclerosis story and mapped to human plaque cell states using only public data.
Core trick
Claude-driven workflow builds a claim-level evidence graph and scores single-cell plaque datasets, with reviewer agents auditing each gate.
Description
PYRIN-PLAQUE Digital Twin investigates whether MEFV/pyrin inflammasome biology can be separated from the saturated NLRP3 atherosclerosis story and mapped to human plaque cell states using only public data.
We built an auditable Claude Science workflow that combines a 57-claim literature evidence graph, verified public datasets, single-cell plaque module scoring, bulk plaque triangulation, a dimensionless pyroptosis ODE, caveat-aware priority scoring, and reviewer-agent audits. The project deliberately separates a pyrin-specific regulatory backbone from NLRP3 and excludes shared downstream genes such as PYCARD, CASP1, GSDMD, IL1B, and IL18 from discriminative scoring.
In GSE159677 single-cell plaque data, MEFV was sparse overall but enriched in macrophage/myeloid cells, and PYRIN_BACKBONE scored highest in the myeloid compartment while NLRP3 localized differently. In GSE120521 bulk plaque data, PYRIN_BACKBONE rose in unstable plaque, but this signal largely collapsed after myeloid-marker adjustment. The ODE model similarly suggested that generic inflammatory priming may dominate pyrin-specific threshold effects.
The result is not a validated drug target or clinical claim. It is a transparent, reproducible, caveat-aware hypothesis engine that prioritizes the MEFV/pyrin threshold axis for future experimental validation while preserving confounded and weak findings instead of hiding them.
README excerpt
# PYRIN-PLAQUE Digital Twin **An auditable Claude Science workflow that separates MEFV/pyrin biology from the saturated NLRP3 atherosclerosis literature and maps a caveated myeloid pyrin-permissiveness hypothesis in public plaque data.** --- ## What this is / what this is not **This is:** - Public-data-only (no private data, no PHI) - Hypothesis-generating - Reproducible (pinned environment, named scripts, checksums) - Auditable (six reviewer-agent gates, full provenance) - Claude Science-driven **This is NOT:** - Validated target discovery - A clinical prediction tool - A drug recommendation - Proof that MEFV causes atherosclerosis - Proof that colchicine acts via pyrin in plaque --- ## Why pyrin, not just NLRP3? The NLRP3 inflammasome in atherosclerosis is **saturated and well-studied** (CANTOS/canakinumab, colchicine trials). Pyrin (encoded by **MEFV**) is a **mechanistically distinct** inflammasome sensor with its own regulatory logic (RhoA/Rac/Cdc42, PKN1/2, 14-3-3, PSTPIP1) — much less mapped in human plaque. To keep the comparison honest, the **shared downstream effectors** (PYCARD, CASP1, GSDMD, IL1B, IL18) are **excluded from discriminative scoring** — the main comparison is PYRIN_BACKBONE vs NLRP3_BACKBONE. The novelty is **deliberately conservative**: pyrin-specific, cell-state-resolved, auditable integration — *not* "first-ever single-cell + bulk + AI" (that combination is common; 6 atherosclerosis papers in our precedent matrix combine ≥3 such methods).