Export

Not a medical test — for research and educational use only

Yeliztli analyses consumer genotyping-array data (23andMe / AncestryDNA), which is not a clinical-grade test. Results are not diagnostic, are not clinically validated, and must not be used to make medical decisions. Array data is especially unreliable for rare, disease-causing variants. Always confirm any finding with an accredited clinical laboratory and discuss it with a qualified clinician or genetic counsellor before acting on it. See the Intended use & disclaimers page for the details and the evidence behind this warning.

Export your data in several formats from the variant table or from Query Builder results:

Format	Description
VCF 4.2	Standard variant call format
TSV	Tab-separated, with all annotation columns
JSON	Structured JSON with nested annotations
CSV	Comma-separated, for spreadsheets
FHIR R4	DiagnosticReport Bundle (JSON) in the FHIR R4 genomics-reporting format, for interoperability with research/genomics tooling — not a clinical diagnostic report (see below). Nuclear variant coordinates are exported in the GRCh37/hg19 reference frame; mitochondrial coordinates use rCRS.

Exports reflect whatever filters or query you have applied, so you can export a focused subset rather than your whole genome.

Report Builder size safeguards

Report Builder keeps interactive exports within fixed resource limits:

• HTML preview and PDF generation accept at most 1,000 reportable findings across the selected report modules.
• FHIR export accepts at most 1,000 Observation resources, and honours the same module checkboxes as the PDF. Selecting fewer modules reduces the bundle; selecting none disables the export, exactly as it disables Preview and Download PDF.

The page disables an action when its selection exceeds the applicable limit. The API repeats the same check and returns HTTP 413 before rendering or building resources, so direct requests cannot bypass the guard. FHIR export also stays disabled when its size cannot be verified or the sample has no annotated variants.

What a module-scoped FHIR bundle contains

The bundle is built from your annotated variants, while the report you curate is built from findings. rsID is what links them, so a scoped export contains the carried variants that a finding in a selected module points at.

A finding that names no variant — a polygenic score, a haplogroup call — contributes no Observation, so a selection made only of those produces a valid but empty bundle, and the page says so before you download it.

Disclosure gates apply to FHIR export, including unscoped exports

A module whose disclosure gate you have not acknowledged is withheld from the FHIR bundle exactly as it is from the PDF — and that holds whether or not you scope the export by module, so requesting an unscoped bundle is not a way around it.

A variant is withheld only when a gated module is its sole reason to appear. One that a non-gated module's finding also names is exported through that module, which is how the PDF path behaves too: it filters findings, not variants.

The FHIR export is not a clinical diagnostic report

The FHIR R4 export produces a DiagnosticReport resource using the standard genomics-reporting format, purely for interoperability with research/genomics tooling. It is research/educational, array-derived, and not clinically validated — it is not a clinical diagnostic report and must not be filed as a clinical result or used to drive clinical decisions. To make this unambiguous to any receiving system, the bundle is marked status: "preliminary" and carries the research-use caveat in the DiagnosticReport.conclusion field. Confirm any finding with an accredited clinical laboratory.

Variant Observation resources use 1-based genomic coordinates and include a LOINC-coded genomic reference-sequence component for the GRCh37/hg19 nuclear build or the rCRS mitochondrial reference, so chromosome positions are not exported as reference-ambiguous values.