Genomics and microbiome intelligence for high-volume programs
Once a program sequences hundreds or thousands of genomes a month, producing data stops being the hard part. Turning it into evidence that is traceable, comparable and analytically sound enough for decisions with clinical, public-health or environmental consequences is.
Verifiable proof
Original files, versions and hashes one click away, with no sign-up.
1
reproducible synthetic cohort
A software demonstration with published inputs, report and hashes. You can run a fresh analysis of these tables in your private workspace.
The check runs in your browser and confirms file integrity. It does not replace scientific review or a fresh run.
60
verifiable synthetic samples
Each row in metadata.tsv describes a generated sample. These are not human participants or clinical observations.
The check runs in your browser and confirms file integrity. It does not replace scientific review or a fresh run.
90
features in the count table
The rows in counts.tsv contain synthetic features used to test the analysis and recover known signals.
The check runs in your browser and confirms file integrity. It does not replace scientific review or a fresh run.
52
public files with SHA-256
The manifest lists the downloads in this frozen example. Each run produces its own artifacts, eligibility states and provenance.
The check runs in your browser and confirms file integrity. It does not replace scientific review or a fresh run.
What held up
A diversity signal holds under the documented depth check.
What did not hold up
Of the two significant contrasts, neither survived the sensitivity test. We publish that too.
Download the synthetic tables and test the analysis ↓ · 0.01 MB · No account, no form and no sales call.
The problem
Thousands of samples produce more results than any team can read with judgement. Without a layer that prioritises and documents, quality depends on who is reviewing that day.
References, versions and annotators drift between deliveries, and technical differences end up mistaken for biological findings.
A result that cannot be rerun or audited is of no use to a clinical committee, a health authority or an external reviewer.
Nexalel turns that flow into evidence: analyses with frozen methods, comparisons that separate signal from artefact, and packages a third party can verify and run again.
What we deliver
Reproduce, compare and substantiate, with public cases you can examine today. Scope and feasibility are agreed for every study before a single file is processed.
Check which results reproduce and document the discrepancies.
A rerun record, a citable package and a method contract with frozen versions.
AvailableComputed today on public microbiome data. In genomics, the public demonstration uses synthetic annotation; real sources are agreed by contract.
Assess what can be compared and which differences prevent a conclusion.
A processed cohort, a declared comparison and an explicit list of what the comparison does not allow you to conclude.
AvailableThe public 16S atlas can be consulted. Comparing a new cohort requires a compatibility assessment and an agreed method; it is not a self-service capability.
A report that connects each conclusion to its evidence and its limits.
A report that separates observation, inference and interpretation, with the intended use stated in writing.
AvailableThe format is published and can be downloaded from the public microbiome case.
Who we serve
National programs, hospitals, laboratories, industry and funders share the same problem with different urgency. Pick the starting point that looks most like yours.
Continuity across cohorts and deliveries.
Traceable evidence for institutional review.
Traceability that travels with every delivery.
Research that partners and advisers can review.
Check whether funded analyses can run again.
How it works
We define the question, the assay, the data rights and the acceptance criteria together.
We check inputs, method and feasibility before anything is processed.
Results, versions, provenance and uncertainty are recorded in the same package.
A dossier another team can verify and run again, together with the record of what did not hold up.
We first agree on the question, the assay, the data rights and the deliverable. Then we decide whether working together makes sense.