The platform
One substrate. Three surfaces.
Everything Genomarker does runs on a single assurance kernel: content-addressed inputs, versioned rules, recorded execution, durable evidence, independent replay. The surfaces differ only in who is asking.
For scientists & research teams
Runtime
Supported computational analyses with methodological assurance built in. Every run produces provenance, a signed evidence package, and a replay path.
- Supported analyses — DE, enrichment, scRNA QC & clustering, proteomics normalization
- Versioned assurance profiles evaluated before execution
- Signed .gmk evidence packages, verifiable offline
- Scheduled independent replays — t+3d, t+1mo, t+1y
For platforms & AI agents
Assurance API
The external system orchestrates; Genomarker constrains, records, and verifies. The same rules, the same evidence, whoever the actor is.
- Schema discovery — agents learn what can be run and how
- Spec validation before execution; blocking findings returned machine-readable
- Idempotent submits, safe to retry and resume
- Actor provenance recorded into every evidence package
For evaluators of scientific claims
Diligence
Service-assisted evidence review of computational claims you did not produce — a verdict report backed by independently inspectable deliverables.
- Claim inventory mapped to analyses and profiles
- Independent replay of every examined claim
- Signed verdict report with structured findings
- Every deliverable verifies offline — the evidence carries the claim
Under every surface
The same lifecycle, the same evidence.
- Ingest
Content-address every input. Identity before analysis.
- Pre-execution
Versioned rules evaluate the spec. Bad science is blocked before it runs.
- Execution
The requested computation runs exactly as specified.
- Evidence
Findings, provenance, environment, results — assembled and signed.
- Preservation
Every replay dependency classified and retained. Graded, not assumed.
- Replay
An isolated environment re-executes and attests to the outcome.
The full reasoning behind the lifecycle lives in the assurance model — Read the assurance model.
Access is reviewed, not sold.
Controlled early access while the public corpus audit completes — research teams, platform partners, and diligence engagements.