NMD-VCell Evidence & Experiment Workbench Audit disease-specific evidence, compare gaps, and freeze the missing test Module: Research Workbench · evidence-to-experiment workflow NMD = neuromuscular disorders
Search Data & API Current release

Current limit: disease modules compare measured states and intervention evidence, but none emits a calibrated patient-level prediction. DMD candidate validation remains prospective.

Research evidence only 21 prioritized candidate perturbation records 2,160 benchmark perturbation targets 250 observed human-myoblast fusion hits Prospective DMD candidate validation pending 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 GSE293514 Stage B0 contributes 250 healthy-human-myoblast fusion hits and 125 individually validated genes No validated DMD perturbation prediction v1.2.0-measured-dmd-evidence Frozen 3 August 2026 Schema 1.1 Model ridge-safe-v2.3 Benchmark repeated-fold-v2.2 Build EA-20260804-49 DOI pending Open evidence boundary →

NMD disease evidence atlas

One platform, four diseases, no forced common model.

DMD remains the candidate-validation core. FSHD, DM1 and SMA now add source-linked patient, causal, correction and intervention-response evidence with disease-specific statistical units and claim ceilings.

Compare evidence axesOpen API manifest
CORE WORKBENCH

DMD

Dystrophin loss

Measured induction/correction and prospective candidate-validation core.

Open disease evidence →FULL_EVIDENCE_MODULE

FSHD

DUX4

Sparse patient-state activation connected to a causal DUX4 time course.

Open disease evidence →FULL_EVIDENCE_MODULE

DM1

DMPK CTG repeat

Repeat excision and patient splicing are linked without collapsing PSI into expression.

Open disease evidence →NEURAL_INTERVENTION_RESPONSE_PLUS_RESIDUAL_MUSCLE

SMA

SMN1 / SMN2

Human neural intervention-response evidence is separated from residual treated-muscle signals.

Open disease evidence →

Shared contract

Evidence states are shared; biology is not flattened.

Architecture diagram for FSHD, DM1 and SMA disease evidence modules with separate anchors, permitted uses and claim ceilings.
Available
  • Disease-specific measured and intervention/correction evidence
  • Typed candidate-establishment logic
  • Downloadable source data and API objects
  • Explicit negative and null results
Still unavailable
  • One common disease score
  • Calibrated cross-disease virtual-cell inference
  • Patient-level simulation
  • Treatment ranking or clinical recommendation