Model Card Audit
External perturbation transfer diagnostic
This card states exactly what the model was allowed to learn, predict, evaluate and claim in this release. Architecture labels and external publications do not create local DMD prediction performance.
Input, output and current result
InputFrozen 55-target external aggregate diagnostic
→OutputDirectional-transfer assessment
- Training context
- HepG2 response substrate
- Held-out task
- External aggregate target set · G1
- Current result
- The current model did not transfer directionally.
- Permitted use
- Method diagnostic; not a cell-population or DMD benchmark
Audit axes
The model is reviewed across twelve required axes.
- model identitydeclared executed unsupported
- training data contextdeclared executed unsupported
- input feature spacedeclared executed unsupported
- output estimanddeclared executed unsupported
- held out task and splitdeclared executed unsupported
- permanent baselinesdeclared executed unsupported
- primary metricsdeclared executed unsupported
- current result or locked statedeclared executed unsupported
- leakage and grouping controlsdeclared executed unsupported
- calibration uncertainty and abstentiondeclared executed unsupported
- permitted and prohibited usesdeclared executed unsupported
- unlock conditiondeclared executed unsupported
Permanent baselines
- zero change
- training-response mean
- ridge
Primary metrics
- aggregate directional-agreement diagnostic
Leakage control
same-task split documented; disease-transfer grouping not applicable
Calibration / OOD
not_calibrated_for_dmd_response · not_validated_for_dmd_muscle_or_patient_context
Claim boundary
What this model card does not authorize
- validated dmd disease response prediction
- therapeutic efficacy or safety prediction
- patient specific or clinical decision support
- performance inherited from external foundation models
Unlock rule: Add harmonized cell-level truth, exact feature alignment and a preregistered external estimator. Automatic unlock is disabled.