Skip to main content

Already have a MyHeritage DNA export?

MyHeritage raw DNA analysis

Editorial revision:

Upload the original CSV, supported tab-delimited TXT or its recognised ZIP and turn usable rsIDs and genotypes into a structured, sourced report. DNA Info Lab supports the observable MyHeritage layouts its parser tests, separates clinical, medication and research evidence, and shows the coverage this file actually provides.

  • ✓ CSV and five-column tab-delimited layouts tested
  • ✓ Original CSV, supported TXT or recognised ZIP
  • ✓ Free findings before any purchase

The MyHeritage CSV and tab-delimited layouts supported

MyHeritage documents a five-column, tab-delimited text export: rsID, chromosome, position, allele 1 and allele 2. DNA Info Lab reads that layout with both alleles and also supports the tested CSV layouts with quoted values and either RESULT or separate allele columns. A .csv filename selects the MyHeritage reader; a MyHeritage header marker, including “##fileformat=MHv1.0”, identifies the source in TXT. Without that marker, a supported five-column tab layout can still be read structurally without confirming MyHeritage as its source.

RSID

The first column has surrounding quotes removed and validated as a public rs identifier or an imported-provider identifier beginning with “i”. The literal header row is not imported.

CHROMOSOME

The chromosome label from the original export is trimmed and stored with the variant.

POSITION

The third field must parse as an integer coordinate. Rows with missing or invalid positions are skipped.

RESULT

In the four-column MHv1.0 layout, RESULT contains the allele pair as one value, for example “AG”.

ALLELE1 / ALLELE2

In the alternative tested header, the parser reads both allele fields and joins them into the stored genotype. This layout is accepted instead of RESULT, not in addition to it.

Blank and partial genotypes are not findings

Rows with an empty genotype, “00”, “--”, a zero within the allele pair or a hyphen within it are skipped. These values describe absent or incomplete calls in the source file; they do not establish that a variant is biologically absent.

A MyHeritage export is selected-marker data

MyHeritage says kits processed before January 2026 were read with a genotyping chip of about 700,000 markers, and that new samples are processed with whole-genome sequencing. The raw-data CSV/TXT is still a table with one variant per row; the sequencing reads of a WGS kit come in a separate CRAM file of about 2 GB, which DNA Info Lab does not read. Neither file is a clinical laboratory result.

DNA Info Lab can only analyse rows present with a usable genotype. When an rsID is missing, the file may never have included it. When a genotype is skipped, the measurement may have been a no-call. When no report item appears, the current filtered catalogue may simply contain no supported record for that row.

The report therefore describes imported coverage and matched evidence. It does not reconstruct variants between markers, infer unmeasured rare variants or turn a blank report section into evidence of low risk.

Quoting, coordinates, build and strand

The parser removes one surrounding pair of double quotes from CSV values, trims line endings and retains the chromosome and coordinate supplied by MyHeritage. It does not infer the genome build, translate coordinates between builds or normalise alleles to another strand.

Catalogue identity is anchored on rsID, so a position is not used by itself to decide which variant a row represents. Allele orientation still affects GWAS dosage and exact PharmGKB genotype matches, which is why an untouched provider export is safer than a spreadsheet-resaved or hand-converted copy.

The parser is intentionally bounded: it supports the known RESULT and split-allele headers. It does not claim that any comma-separated genetics file, exported report PDF or third-party conversion is a MyHeritage raw-data file.

Three evidence sources, three meanings

Imported genotypes can overlap with three catalogues, but DNA Info Lab never presents those overlaps as interchangeable claims.

01

ClinVar

An rsID overlap can surface imported ClinVar condition records, classification, review status and provenance. A consumer-test call that may matter clinically must be confirmed with an appropriate accredited test.

02

PharmGKB

The report shows an imported medication annotation only when both rsID and genotype match. It explains evidence about published drug response and never tells a person to start, stop or change treatment.

03

GWAS Catalog

Research associations can contribute when the imported genotype contains the catalogue effect allele. Traits, effect units, studies and population limitations remain visible rather than being collapsed into a personal diagnosis.

Why two MyHeritage exports can yield different reports

MyHeritage says its previous chip and its current whole-genome sequencing look at different sets of markers, which changes the raw-data file. DNA Info Lab calculates report coverage from the usable rsIDs actually imported; it does not promise a fixed number of markers or findings for the MyHeritage brand.

Source coverage changes too. A valid genotype can be absent from the filtered ClinVar, PharmGKB or GWAS datasets currently served, while a single physical rsID can carry several study associations or clinical condition records. The report labels positions, studies, variants and medicines in their correct units.

Provider differences are real: a MyHeritage file may overlap differently with the catalogues than a 23andMe or AncestryDNA file. An area with no supported matches is retained as an honest coverage result rather than filled with marginal evidence.

Inspect the finished report before uploading

The public demo uses synthetic genotypes and the same report components as production. It shows the ClinVar, pharmacogenomics and GWAS sections, simple and scientific reading modes, citations and premium boundary without publishing anyone’s MyHeritage data.

Open the synthetic report →

Use the original MyHeritage download

Request the DNA data file from Manage DNA kits. MyHeritage emails a download link that is valid for 24 hours. Do not upload a health report, an ethnicity PDF or a hand-edited copy: those are different artefacts from the raw-data file.

How to download MyHeritage raw DNA data →

The original raw-data file is parsed, then discarded

The original MyHeritage upload is parsed in server memory and discarded. Extracted variants and derived report results remain on the service’s EU infrastructure so the account can return to the analysis, subject to account deletion and the applicable retention rules.

  • ✓Genetic data is not sold and genotype, condition, medicine and report content are excluded from advertising analytics.
  • ✓Optional AI explanations run only after an explicit request and use bounded report context rather than the original raw-data file.
  • ✓The account owner can export stored account data or delete the account through product controls.
Read the privacy policy →

Limits to keep beside every MyHeritage result

  • —DNA Info Lab is educational software, not a diagnosis, clinical laboratory, treatment recommendation or MyHeritage service.
  • —Consumer-test calls can be wrong; clinically important matches require confirmation by an appropriate accredited assay.
  • —A variant absent from the report may be untested, a no-call, filtered from the current catalogue or outside the evidence sources entirely.
  • —GWAS estimates describe studied populations and may not transfer precisely across ancestry, environment, age, sex or individual context.
  • —The analysis does not calculate ethnicity percentages, find relatives, reconstruct family trees or access a MyHeritage account.

MyHeritage raw DNA questions

What MyHeritage file type can I upload?

Use the original .csv export, the supported five-column tab-delimited .txt export, or a recognised ZIP containing one such file. A .txt extension alone does not establish compatibility. CRAM sequencing reads are not supported. The parser does not accept an interpreted PDF or screenshot as genotype data.

Which MyHeritage columns are supported in CSV or TXT?

RSID, CHROMOSOME and POSITION are required. The genotype must be in RESULT, or in both ALLELE1 and ALLELE2 when that alternative header is present.

Does the parser handle quoted CSV values?

Yes. It trims values and removes one surrounding pair of double quotes before validating identifiers, coordinates and genotypes.

What happens to 00, -- or a partial allele?

The row is skipped. Zero, hyphen and blank forms are treated as missing or incomplete calls, never as biological alleles or negative findings.

Is an MHv1.0 line enough to identify any file?

A MyHeritage header marker, including “##fileformat=MHv1.0”, selects the MyHeritage reader for TXT; a .csv filename also selects it. Identification does not guarantee usable rows: they must still follow a supported layout. A generic TXT file is not automatically compatible.

Does DNA Info Lab detect the genome build or flip strands?

No. It retains source coordinates, matches identity by rsID and does not infer builds, lift coordinates or rewrite allele orientation.

Will it reproduce MyHeritage ethnicity or relative matches?

No. DNA Info Lab cross-references genotypes with scientific evidence sources. It does not calculate ancestry percentages or search genealogical databases.

Can I see anything before paying?

Yes. Processing and the defined free findings come before the optional one-time purchase of the complete report.

Read the evidence your MyHeritage file supports

Start with the untouched export, inspect its real coverage and free findings, and choose the complete sourced report only if it is useful to you.

Analyse my MyHeritage file →