STRC literature-params — cross-hypothesis parameter DB
Single source of truth for every numeric input used in models/. Each row: value, units, primary source, page/figure, used-in-hypotheses. Pull from here; do not hardcode from memory.
Why this exists
2026-04-23 audit found ~12 hardcoded constants in hydrogel_phase4d*.py / phase4e*.py / phase1*.py claiming literature backing without corresponding PDFs in papers/. Fixing before Holt/Zhang/Barr-Gillespie reviewer outreach.
Topic files
| Topic | File | Domain agent | Status |
|---|---|---|---|
| Cochlear PK/PD | cochlear-pkpd | Domain 1 (Sonnet) | ✅ 6 papers retrieved |
| Actin + WH2 kinetics | actin-kinetics | Domain 2 (Sonnet) | ✅ 7 papers retrieved |
| Stereocilia biomechanics | stereocilia-mechanics | Domain 3 (Sonnet) | ✅ 4 papers retrieved |
| RADA16 / SAP geometry | rada16-geometry | Domain 4 (Sonnet) | ✅ 6 papers retrieved |
| STRC-TMEM145 interactions | strc-tmem145-interactions | Domain 3 → aggregated | ✅ no-Kd gap documented |
| Calcium oscillation kinetics | calcium-oscillation | h05 audit (Sonnet) | ✅ 7 papers retrieved; 3 phantoms caught |
| Pharmacochaperone h01 | pharmacochaperone | h01 audit (Sonnet 4.6) | ✅ audited 2026-04-23; 0 phantom cites; 2 MEDIUM inconsistencies |
| Piezoelectric materials | piezoelectric-materials | h02 audit (Sonnet 4.6) | ⚠ manufacturer specs only; 3 primary papers needed |
| Tectorial membrane mechanics | tectorial-membrane | h02 audit (Sonnet 4.6) | ⚠ 5 papers needed; mech mismatch flag added |
| Avidity, interface area, AF3 calibration | avidity-and-dimers | h26 audit (Sonnet 4.6) | ✅ audited 2026-04-23; Ceff/Kd theory documented; ipTM≠Kd confirmed; 3 phantom Kd claims flagged in h26 |
Cross-hypothesis reuse matrix
| Parameter | Hypothesis users |
|---|---|
| Cochlear fluid volumes | h01, h03, h04, h05, h06, h09 |
| RWM permeability by MW | h03, h04, h06, h09 |
| Stereocilia actin geometry | h02, h03, h09, h26 |
| WT bundle stiffness | h02, h03, h09, h26 |
| STRC × TMEM145 Kd | h03, h09, h26 |
| WH2 × G-actin kinetics | h09 (unique) |
| RADA16 fibril geometry | h09 (unique) |
| Peptide proteolysis, perilymph | h06, h09 |
| AC1 (ADCY1) Ca/CaM kinetics | h05 (unique) |
| CaMKII / calcineurin kinetics | h05 (unique) |
| CREB-P dephosphorylation t½ | h05 (unique) |
| MET channel g, f_Ca, n_channels | h05; h02 (piezo) |
| Avidity C_eff for protein homodimer | h26 |
Update protocol
- Never hardcode a literature value in a model script. Pull from a paper note in
papers/via docstring reference. - When a new paper lands: append row to the relevant topic file + update paper note’s
## Numbers that mattersection + update this hub’s timestamp. - When a parameter changes (better paper supersedes): mark old row
[superseded]— keep for replication.
Audit links
- _audit-2026-04-23 — the originating audit that triggered this DB
- STRC Hydrogel Phase 4 Computational Campaign — primary consumer
- STRC Computational Scripts Inventory — script-level tracking
Connections
[part-of]STRC Hypothesis Ranking[see-also]STRC Computational Scripts Inventory[about]Misha