NMD-VCell Disease Virtual Cell Platform Specify a cell-state question, inspect computability, and design the missing test Module: Predict · evidence-to-experiment workflow NMD = neuromuscular disorders

Current limit: the site can compare existing evidence, but no candidate has yet been independently repeated in a DMD muscle model.

Research evidence only 21 prioritized candidate perturbation records 2,160 benchmark perturbation targets 0 independent DMD replications Boundary & release
21 Workbench candidates are surfaced for consideration, not ranked as validated targets 2,160 target-level perturbation responses form the frozen HepG2 benchmark substrate No validated DMD perturbation prediction v1.0.0-database-resource Frozen 25 Jul 2026 Schema 1.1 Model ridge-safe-v2.3 Benchmark repeated-fold-v2.2 Build EA-20260730-42 DOI pending Open evidence boundary →

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.

NMDVCELL-TRANSFER-DIAGNOSTIC-1.0executed but current result unsupported

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.

No scalar model score
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

Unlock rule: Add harmonized cell-level truth, exact feature alignment and a preregistered external estimator. Automatic unlock is disabled.