DMD + matched control
DMD-derived human myogenic system plus matched healthy or isogenic-corrected system.
Flagship 01 · draft protocol
DMD Minimum Perturbome
A 24-slot prospective pilot designed to produce the first independently replicated DMD functional perturbation outcomes.
Important. Twenty-four selection slots are defined; the genes, primary endpoint, effect margin, sample size and QC thresholds remain deliberately unfrozen. This is not a registered experiment yet.
Selection frame
Do not select only the highest-scoring candidates
| Stratum | Slots | Purpose |
|---|---|---|
| Muscle-context candidates | 6 | Highest immediate disease-context utility |
| High uncertainty | 4 | Calibrate abstention and active learning |
| Mechanism diversity | 4 | Avoid a single-pathway panel |
| Known myogenesis/DMD controls | 4 | Assay-positive calibration |
| Random or negative controls | 4 | Estimate selection enrichment |
| Toxicity controls | 2 | Calibrate the safety boundary |
Factorial axes
Every outcome retains context, state, time and replication
Direction and modality explicit
CRISPRi, CRISPRa, KO or siRNA chosen per question; two independent reagents where feasible.
Myoblast → myotube
Proliferation, early differentiation, fusion and maturing myotube are not pooled into one context label.
Reagent · donor · batch
Cells inside one well are measurements, not independent biological replicates.
Mechanism timeline
Early mechanism precedes late phenotype
T0Baseline
Control state and assay eligibility.
6–12 hMechanism
Direct molecular and signalling response.
24–48 hTransition
Regulatory program and state proportion.
4–7 dFunction
Fusion, morphology, DMD function and toxicity.
Endpoint hierarchy
Target engagement → muscle function → DMD function → safety cost
Planning primary endpoint
Fusion index at an assay-calibrated late timepoint; must be confirmed or replaced before registration.
The final primary endpoint and timepoint must be immutable before outcome access.
Primary estimand
The DMD-versus-matched-control difference in perturbation effect on the frozen primary functional endpoint, with reagent-specific and donor-specific estimates retained.
DMD functional panel
membrane integrity · contraction · calcium handling · mitochondrial function · oxidative stress · injury susceptibility · regeneration response
Safety and cost
viability · proliferation · differentiation blockade · global stress · essentiality · off-target phenotype
Outcome ontology
Failure, null and heterogeneity are first-class results
robust_responderdmd_specific_responderdonor_variablestate_specifictoxic_respondernull_with_equivalence_margindiscordantqc_failureinconclusive
Hard gates
What must be frozen before the first experiment starts
- freeze the 24 slot assignments and selection rules
- freeze one primary endpoint and numeric minimally important effect
- freeze negligible-effect margin
- freeze sample size or pilot variance rule
- freeze target-engagement, viability and imaging QC thresholds
- confirm biospecimen, institutional and laboratory approvals
- register the immutable Study and Prediction objects before outcome access
Boundary. This is a draft experiment-registration contract. It contains no assigned 24-gene panel, no experiment, no outcome and no efficacy claim.