STRC h05 Ca Oscillation — Parameter Provenance Audit 2026-04-25

Second audit extending STRC h05 Parameter Provenance Audit 2026-04-23. The 2026-04-23 audit removed 4 phantom citations from scripts and flagged ⚠ rows. This pass completes the literature retrieval for all retrievable parameters, identifies one additional phantom (PMID 21458670 for CaMKII), confirms 3 STRC parameters as permanently unmeasured, and creates 7 paper notes.


PHANTOM resolution — high priority (3 + 1 bonus)

PHANTOM 1: “Sharma 2018” — STRC mRNA t½

Resolution: NOT MEASURED — value retracted from model as primary measurement.

Exhaustive literature search finds no paper by any author named Sharma from 2018 measuring STRC mRNA stability or half-life. No primary paper measuring STRC mRNA t½ in any cell type exists in PubMed or bioRxiv as of 2026-04-25. The parameter STRC_MRNA_DECAY_S = 1e-4 (t½ ~2 h) and k_mRNA_deg = 0.0004 (t½ ~30 min) in the two scripts are both unsourced calibration parameters. Scripts already carry # ⚠ UNSOURCED comments from 2026-04-23 audit; no additional script changes needed. Literature table updated to reflect “NOT MEASURED.”

PHANTOM 2: STRC protein t½ — no source

Resolution: NOT MEASURED — value retracted from model as primary measurement.

No primary paper measures STRC protein half-life in OHCs or any mammalian system. The 38 h (pivot) vs 30-day (rbm24) values are 20× apart. Both are unsourced. No proteomics paper with STRC turnover data found. Scripts already carry # ⚠ UNSOURCED / # ⚠ fitted comments. Literature table updated.

PHANTOM 3: “Krey 2015” — STRC copy number 15,000/OHC

Resolution: NOT MEASURED — value retracted from model as primary measurement.

Krey et al. 2015 is Wilmarth et al. Scientific Data (PMID 25977827) — stereocilia proteome from mouse cochlea. STRC is not quantified in that dataset. No published proteomics paper (including mass spectrometry surveys of stereocilia) quantifies STRC copy number per OHC. The target_protein = 15,000 is a model calibration set-point with no empirical basis. Scripts carry # ⚠ UNSOURCED comment from 2026-04-23. Literature table updated.

BONUS PHANTOM 4: “Chao 2011 Cell PMID 21458670” — CaMKII

Resolution: PHANTOM — PMID 21458670 is a completely different paper.

PMID 21458670 resolves to: Edlich F et al. “Bcl-x(L) retrotranslocates Bax from the mitochondria into the cytosol.” Cell 2011;145(1):104-16. This is an apoptosis paper with zero CaMKII content. The correct CaMKII paper is Chao LH et al. “A mechanism for tunable autoinhibition in the structure of a human Ca2+/calmodulin-dependent kinase II holoenzyme.” Cell 2011;146(5):732-45. PMID: 21884935. Paper note created: 2011-chao-camkii-holoenzyme-structure. Scripts corrected in ca_oscillation_rbm24_ode.py.

New finding from Chao 2011 (PMID 21884935): CaM EC50 for full holoenzyme activation = 6.0 µM (short linker, Hill n=2.1). Model Kd_CaMKII = 1 µM is ~6× lower — a model choice, not a measurement for the OHC isoform. Flagged in scripts.


Additional citation corrections (not full phantoms but wrong papers)

“Mammano 1999 Pflugers Arch” → CORRECTED

Mammano 1999 (PMID 10436049, J Neurosci) = “ATP-Induced Ca2+ release in cochlear outer hair cells: localization of an IP3-gated Ca2+ store.” This is not a PMCA kinetics paper. The correct sources for k_extrusion are:

  • Yamoah et al. 1998 J Neurosci (PMID 9425003) — PMCA density + functional necessity
  • Beurg et al. 2010 J Neurophysiol (PMID 20427623) — max extrusion 3.7 fmol/s, τ ~50 ms

Paper notes created for both. Scripts updated.

”Müller et al. 2002” buffer ratio → CORRECTED

No Müller 2002 paper on OHC calcium buffering found in any search. The correct primary source is Hackney et al. 2005 J Neurosci (PMID 16120789): oncomodulin 2-3 mM + calbindin 15-230 µM in rat OHC = ~5-6 mM total Ca2+-binding sites; OHC buffer is 110× that of IHC. Paper note created: 2005-hackney-calcium-buffering-proteins.

”Houslay 2010” PDE4 kinetics → CORRECTED

No Houslay primary paper from 2010 on PDE4 kinetics identified. Closest: Houslay & Adams 2003 Biochem J (PMID 12444918) — a review with no original kinetics. For primary Km measurement: Owens et al. 1997 Biochem J (PMID 9337850): Km = 4.8 µM for full-length PDE4A. Model PDE4_KM_NM = 4000 nM (4 µM) is consistent. PDE4 Vmax remains a model-fitted parameter with no primary source. Paper notes created for both.

”Lumpkin 2001” wrong year → CORRECTED

Paper is from 1998: Lumpkin & Hudspeth J Neurosci 1998 (PMID 9698322). Scripts updated in both models.

”Fettiplace 2017” wrong year in n_channels comment → CORRECTED

Fettiplace & Kim 2014 Physiol Rev (PMID 24987009). n_channels ~134 is consistent with 2 channels per tip link × 60-100 stereocilia per OHC (per Fettiplace 2014). Script updated.


⚠ rows converted to ✅ this pass

RowParameterNew status
Apical compartment volumeV_apex ~50 fL✅ Lumpkin & Hudspeth 1998 PMID 9698322
Resting [Ca2+]20-310 nM✅ Lumpkin & Hudspeth 1998 PMID 9698322
Ca2+ buffer capacity100-1000×✅ Hackney 2005 PMID 16120789
PMCA extrusion ratek ~30 /s✅ Yamoah 1998 PMID 9425003 + Beurg 2010 PMID 20427623 (as model estimate)
PDE4 Km~4 µM✅ Owens 1997 PMID 9337850 (4.8 µM direct measurement)
CaMKII EC50/Kd~6 µM✅ Chao 2011 PMID 21884935 (EC50 6 µM for short-linker holoenzyme)

Paper notes created (7 total this pass)

  1. 1998-lumpkin-hudspeth-stereocilia-calcium-regulation (PMID 9698322)
  2. 1998-yamoah-pmca-stereocilia-extrusion (PMID 9425003)
  3. 2010-beurg-calcium-balance-ohc (PMID 20427623)
  4. 2005-hackney-calcium-buffering-proteins (PMID 16120789)
  5. 2011-chao-camkii-holoenzyme-structure (PMID 21884935)
  6. 2003-houslay-adams-pde4-review (PMID 12444918)
  7. 1997-owens-pde4a-kinetics (PMID 9337850)

Remaining ⚠ / ❌ blockers

ParameterStatusReason
STRC mRNA t½⚠ NOT MEASUREDNo primary literature exists. Scripts carry WARNING comment.
STRC protein t½⚠ NOT MEASUREDNo primary literature exists. Cross-script inconsistency remains.
STRC copy number/OHC⚠ NOT MEASUREDNo proteomics paper quantifies STRC/OHC.
AC1 Vmax (cell-normalized)⚠ model choice30-60× above membrane-prep value; justified as volume normalisation
PDE4 Vmax⚠ no primaryFitted model parameter; no OHC-specific source
CaMKII k_auto⚠ not measuredFitted; Chao 2011 is structural only
CREB dephos k⚠ mismatch2-4× faster than Gonzalez & Montminy 1989; no OHC source
CRE half-saturation K_half⚠ unverified”Cha 2010” PMID not confirmed; no alternative found
k_mRNA_deg (STRC)⚠ 4× cross-script mismatchBoth unsourced; no primary data
k_prot_deg (STRC)⚠ 20× cross-script mismatchBoth unsourced; no primary data

Wet-lab gates needed to resolve STRC kinetic parameters:

  • STRC mRNA stability: actinomycin D chase in OHC line (HEI-OC1 or primary mouse OHC)
  • STRC protein t½: cycloheximide chase + immunofluorescence in OHC
  • STRC copy number: single-cell proteomics (SciPlex or similar) in sorted OHC

Model-script corrections summary

FileLines changedWhat
ca_oscillation_rbm24_ode.pyn_channels comment”Fettiplace 2017” → “Fettiplace & Kim 2014 PMID 24987009”
ca_oscillation_rbm24_ode.pyf_Ca comment”Lumpkin 2001” → “Beurg 2006”
ca_oscillation_rbm24_ode.pybuffer_ratio commentAdded “Hackney 2005”; removed phantom “Müller 2002”
ca_oscillation_rbm24_ode.pyV_apex comment”Lumpkin 2001” → “Lumpkin & Hudspeth 1998 PMID 9698322”
ca_oscillation_rbm24_ode.pyk_extrusion commentCorrected “Mammano 1999” → “Yamoah 1998 + Beurg 2010”
ca_oscillation_rbm24_ode.pyKd_CaMKII blockPMID 21458670 (phantom) → PMID 21884935; added WARNING
ca_oscillation_ac1_creb_pivot.pyK_EXTRUSION_S commentCorrected “Mammano 1999 Pflugers Arch” attribution
ca_oscillation_ac1_creb_pivot.pyV_APEX_L comment”Lumpkin 2001” → “Lumpkin & Hudspeth 1998 PMID 9698322”
ca_oscillation_ac1_creb_pivot.pyPDE4 comment”Houslay 2010” phantom → “Owens 1997 PMID 9337850”; added WARNING

lit_audit state

lit_audit: partial (see blockers above)

Rationale for partial not fixed:

  • 3 STRC parameters (mRNA t½, protein t½, copy number) have NO primary literature. Cannot be resolved by literature retrieval — require wet-lab measurement.
  • “Cha 2010” CRE half-saturation PMID remains unconfirmed.
  • PDE4 Vmax and CaMKII k_auto remain model-fitted.

fixed would be premature. These are fundamental model parameters with no empirical basis; they are retracted from primary-measurement status and documented as model calibration set-points.


Ranking delta

AxisPre-auditPost-auditReason
Mech33Topological result unchanged; phantom resolution did not change mechanism sketch
Deliv44Acoustic delivery unchanged
Misha-fit22Unchanged

Tier: A (unchanged). This audit closes known phantom debt and improves citation integrity. The 3 STRC kinetic parameters remain unmeasured — this is now explicit and documented, not hidden. The hypothesis’s scientific weakness remains Misha-fit (2), not computational fidelity.


Connections