Question answeredWhat is draft, frozen, registered, running, measured, released, superseded or intentionally absent?
Scope & evidence unitTwo layers are explicit: the DMD prospective lifecycle and the released four-disease evidence contract.One immutable versioned research object; a missing record is never interpreted as a negative result.
ProducesLifecycle states for comparisons, Study Cards, protocols, ModelRuns, predictions and outcomes, including honest zero states.
Cannot establishThe public registry is read-only; it does not accept uploads, create persistence or promote zero outcomes into validation.
Next handoffContribute a traceable outcome or evidence object, then apply the formal transition rules. Continue →
Product contract · WORKFLOW-OUTCOME-CARD-1.0
Outcome Card: What was registered, measured or failed—and what conclusion changed?
Study identityFrozen prediction when usedMeasured outcomePositive · null · toxic · failedProvenanceEvidence transition
Two prospective lifecycles
Truth generation and prediction validation are different scientific jobs.
A truth-generating experiment creates the evidence needed to train or calibrate a model. A prediction-validation experiment tests an already eligible model prospectively.
Governance ruleTruth-generation studies require an immutable registered Study before execution but do not require a Prediction. Prediction-validation studies require both an eligible ModelRun and a frozen Prediction before outcome access.
Null, toxic, discordant and inconclusive outcomes are first-class records and are not removed from the lifecycle.
Decision Cards connect uncertainty to a result that can change the record
These cards are editable drafts. They organize a discriminating experiment, an evidence audit and a prospective panel freeze without claiming a winner.
PROTOCOL:v1.2.0-measured-dmd-evidence:DMD-MIN-PERTURBOME:1.1-PANEL-DRAFT provisionally assigns 14 candidate genes, 10 calibration controls and six prospectively sealed roles. It remains editable until panel signoff, assay calibration, approvals, reagent identities and an immutable Study object are complete. A Prediction object is required only if the study later enters the prediction-validation track.
State authority:PANEL_DRAFT_ASSIGNED_SIGNOFF_PENDING · effective 2026-08-06 · next PANEL_FROZEN_ENDPOINT_FROZEN_READY_FOR_REGISTRATION.
REGISTEREDImmutable; timestamp and checksum required
MEASUREDOutcome linked without overwriting prediction
SUPERSEDEDNew revision links to prior object
Current boundary. This static public beta exposes read-only contracts and draft objects. It does not accept uploads or claim registry persistence without an approved storage, identity and governance policy.
Released governance object
PredictionReadiness gate
PREAD:v1.2.0-measured-dmd-evidence:EA-20260817-57 is a released governance object. It states that P0 infrastructure is available, P1 research-use DMD prediction is locked pending independent disease-relevant perturbation validation, and P2 clinical prediction is outside the current release.
State-transition program and DMD Minimum Perturbome
PROGRAM:v1.2.0-measured-dmd-evidence:STATE-TRANSITION:1.0 defines the dual-track scientific program. The no-new-wet-lab pack contributes eleven bounded evaluation modules, including a partial UniPert source-and-identifier feasibility audit; the 24-slot Minimum Perturbome has 14 candidate and 10 control assignments and is awaiting study registration.
HMPN:v1.2.0-measured-dmd-evidence:EA-20260817-57:STAGE-B0 freezes 92 evidence-typed edges from GSE293514, including 31 exact fusion-screen effects, 17 individual-validation edges and two qualitative grouped CROP-seq state edges. Its 21-candidate audit records 9 assessed genes, 0 fusion hits and 12 genes outside the library. DMD prediction edges remain 0.
V1ReleaseGate, ExternalAuthorizationDossier and PredictionRegistry
V1GATE:v1.2.0-measured-dmd-evidence:EA-20260817-57 records formal database-paper readiness without authorizing formal v1.0. AUTHD:v1.2.0-measured-dmd-evidence:EA-20260817-57 records the DOI/licence/maintainer dossier that remains pending external input. PREG:v1.2.0-measured-dmd-evidence:EA-20260817-57 records the empty prospective prediction registry; no prediction or measured-outcome record has been registered yet.