Research readiness workbench | NMD-VCell Research Workbench

Research Workbench

Move a DMD state-transition hypothesis through registration and outcome.

Search and comparison remain inputs. The research unit is perturbation × state × disease context × time × phenotype, with reagent, donor, toxicity and null outcomes retained.

Evidence ceilingL2/L3a today · no independent DMD perturbation replication · no therapeutic or clinical validationClaim boundary

Open scientific program Open 24-slot pilot Create Study Card

Bioinformatics result snapshots

Results you can read before opening the ledgers

One data-driven SVG turns the frozen release objects into a visual summary: object scale, candidate action queue, evidence-depth funnel, DMD source-agreement strip and benchmark gates. It is descriptive, not a ranking or prediction claim.

Download result SVG

Bioinformatics result snapshot figure summarising NMD-VCell object scale, candidate action queue, evidence depth, DMD source agreement and benchmark gates. — Static SVG generated from release manifests, candidate records and strict outcome objects. Gene symbols, API fields and numeric gates remain unchanged; no composite score or target ranking is generated.

Open WorkbenchInspect readiness and action queue Open 21 hypothesesFilter current actions and gaps Open BenchmarkAudit strict 0/16 gates

Quick actions

Move from a gene list to an auditable object

No login · local workflow

01Search one geneOpen a typed record, evidence level, DMD state and gap badge. 02Compare a gene listRender a matrix while preserving mapping provenance and input order. 03Create a Study CardGenerate a bounded next-experiment scaffold without ranking. 04Register a state-transition testBind context, state, time, function, toxicity and replication before outcome access.

Open object registry to inspect lifecycle states, released comparison objects and draft Study Cards.

Data reality

The scientific bottleneck is visible as a funnel

Evidence gaps

21L2 observationsHepG2 perturbation observations

↓ 43%

9L3a screensExternal context, not replication

↓ 0%

0L3b replicationsIndependent myogenic perturbation absent

↓

0L4 validationsDMD functional validation absent

↓

0L5 validationsTherapeutic / clinical out of scope

Multi-fidelity path

F0–F4 shows what must happen next

F3/F4 not crossed

F0Resource17,921

Gene records and 120 SQLite tables.

F1HepG221/21

Observed same-context perturbations.

F2Context9/21

L3a screens, not replication.

×evidence break F3Independent0/21

Independent perturbation missing.

F4DMD function0/21

No current record has crossed F3 or F4.

Current scientific breakpoint: F2 → F3. The next experimental resource should produce independent disease-relevant perturbation replication before any prediction claim expands.

Capability matrix

Active tools are separated from scientifically locked tools

Prediction gate

Module	Data condition	Technical state	Scientific state	User action
Evidence Explorer	Available gene and candidate records	Active	L1–L3a evidence inspection	Explore
State-transition program	Released program contract; no experimental outcomes	Registration ready	Dual-track research frame with explicit hard gates	Open
Study Card	21 draft objects	Active	Draft only; prospective registration required	Create
Prediction Card	Schema and empty registry available	Registration ready	No predictions registered in this release	Inspect
Perturbation Lab	Observed HepG2 CRISPRi plus bounded additive pair contract	Preview active	SIM-L0/L2 only; no interaction, KO, OE or fate prediction	Open lab
Counterfactual Explorer	Association evidence is not intervention-identifiable	Locked	Causal claim not authorized	View boundary
Active Learning Queue	Calibration and acquisition budget absent	Locked	Uncertainty not prospectively validated	View registry

System status

Releasev1.0.0-database-resourceDOI pending

Data breadth17,921 genes120 non-empty tables

Predictions0 registeredP0 infrastructure only

External gatesPendingDOI, licence, maintainer metadata

Prediction readiness

P0PASSPrediction infrastructure only

P1LOCKED_PENDING_EXTERNAL_VALIDATIONResearch-use DMD perturbation-response prediction

P2LOCKED_NOT_IN_SCOPEClinical disease prediction or decision support

No P1 or P2 prediction claim can be made until independent disease-relevant perturbation data, prospective registration, external test results, calibration, uncertainty, OOD and abstention analyses are frozen.

Primary next asset

DMD Minimum Perturbome

A compact, well-controlled disease-relevant perturbation panel must precede any larger virtual-cell claim.

DMD and matched control myogenic systems
Two independent reagents where feasible
Imaging-first functional phenotype
Timestamped prospective holdout

Open the draft 24-slot protocol