# NMD-VCell no-new-wet-lab evaluation pack

This package uses released data, public data and new computation to stress-test, falsify, downgrade and prioritize hypotheses before any new wet-lab study.

## Included objects

- `api/v1.1/no_wet_lab_evaluation.json`: authoritative module and claim contract.
- `downloads/no_wet_lab_evaluation_module_registry.tsv`: populated module registry.
- `downloads/no_wet_lab_evaluation_scorecard_template.tsv`: one-row-per-estimate run template.
- `api/v1.1/unipert_feasibility.json`: source, implementation, identifier and evaluation-stage contract for NWLE-10.
- `downloads/unipert_candidate_identifier_preflight.tsv`: 21-candidate identifier preflight; not an encoder result.
- `downloads/unipert_feasibility_scorecard_template.tsv`: one-row-per-estimate UniPert evaluation template.
- `downloads/UNIPERT_FEASIBILITY_PROTOCOL.md`: leakage-safe execution and adoption protocol.

## Required run order

1. Freeze the eligible data universe and identifiers.
2. Freeze the independent unit, leakage-safe split and baseline set.
3. Freeze metrics, uncertainty and multiplicity rules.
4. Execute without outcome-informed changes.
5. Publish estimates, intervals, nulls, failures and artifact URIs.
6. End with KEEP_FOR_COMPUTATIONAL_TRIAGE, DOWNGRADE, ABSTAIN or ESCALATE_TO_EXPERIMENT.

## Claim ceiling

The pack can support reproducibility, same-context computational fit, external compatibility or falsification, and experiment prioritization. It cannot establish DMD perturbation efficacy, causal function, safety or treatment utility.

No new wet lab does not mean no new analysis or no new public-data ingestion. All additions require provenance and a new frozen Scorecard.