Model Card Audit
Disease-conditioned state-transition model
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
InputRequired matched DMD/control myogenic perturbations with donor, state, time and function
→OutputDesired molecular response, state transition, functional effect, toxicity and calibrated uncertainty
- Training context
- No qualifying disease-relevant perturbation training set
- Held-out task
- Planned unseen donor · state · laboratory · disease line · G2–G6
- Current result
- No calibrated DMD state-transition prediction is emitted.
- Permitted use
- Architecture and evaluation contract only
Audit axes
The model is reviewed across twelve required axes.
- model identitydeclared locked no current result
- training data contextdeclared locked no current result
- input feature spacedeclared locked no current result
- output estimanddeclared locked no current result
- held out task and splitdeclared locked no current result
- permanent baselinesdeclared locked no current result
- primary metricsdeclared locked no current result
- current result or locked statedeclared locked no current result
- leakage and grouping controlsdeclared locked no current result
- calibration uncertainty and abstentiondeclared locked no current result
- permitted and prohibited usesdeclared locked no current result
- unlock conditiondeclared locked no current result
Permanent baselines
- zero change
- train mean
- ridge
- nearest-neighbour transfer
Primary metrics
- DES
- PDS
- MAE
- calibration
- AUPRC
- hit rate
- replication
- abstention
Leakage control
registered_requirement_no_executed_split
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: Return independent DMD perturbation outcomes through frozen Study, Prediction and Outcome objects. Automatic unlock is disabled.