Choosing a DNA test: all your options compared
Consumer chip arrays, whole-genome sequencing, Oxford Nanopore long reads — every provider, what each unlocks in Haeckel, how to download your raw file, and a decision guide based on your goals and budget.
If you already have results from a consumer DNA company, the first and cheapest step is to download your raw data file and upload that. Millions of people already have 23andMe or AncestryDNA results they have never done anything with beyond reading the ancestry estimate the company provides. Haeckel can run every analyser — ancestry, haplogroups, health variants, pharmacogenomics, archaic ancestry, polygenic risk scores — on that existing file at no additional cost beyond what you originally paid for the test. Check that option first before ordering anything new.
Type 1 — Consumer SNP chip arrays
SNP chip arrays sequence roughly 650,000 to 900,000 positions across your genome. They are not sequencing the full genome — they are interrogating a fixed set of known-variant positions. For most of Haeckel's analyses this is sufficient: the ancestry informative markers, health-risk SNPs in ClinVar, pharmacogenomic star-allele positions, APOE, and the polygenic score weight sets are all designed around the variants chips reliably genotype. The tests cost $79–200, arrive by post, and return results in three to six weeks.
The most common consumer DNA test. The current v5 chip assays roughly 650,000 SNPs on the Illumina Global Screening Array. Raw data downloads as a tab-separated .txt file inside a zip archive (rsid, chromosome, position, genotype columns). The format is what Haeckel's parser was originally built around, so it is the most reliable path. The Health + Ancestry tier adds BRCA1/2 screening and pharmacogenomic reports from 23andMe, but Haeckel recomputes all of those from the raw file independently — you do not need the upgraded tier to get full Haeckel analysis. One known quirk: the v5 chip uses internal IDs (e.g. i3003137) for some positions that do not have rsIDs assigned, including the two canonical APOE positions. Haeckel's APOE analyser handles these via its 4-tier fallback specifically to accommodate v5 chip files.
Ancestry's chip assays roughly 700,000 SNPs. Raw data downloads as a comma-separated archive with columns rsid, chromosome, position, allele1, allele2 — a slightly different schema from 23andMe that the parser handles automatically. AncestryDNA has the largest consumer DNA database in the world (over 22 million profiles), which benefits their relative-matching service but has no effect on Haeckel analysis since Haeckel works from your genotypes directly. The raw data download requires a two-step email confirmation and may take up to 24 hours to prepare. File size is typically 28–35 MB compressed.
The most affordable major chip test. Uses the same Illumina GSA platform as AncestryDNA at roughly 700,000 SNPs. Raw data is a CSV download from the DNA Tools section of their dashboard. Compatible with all Haeckel analysers. MyHeritage has a smaller relative-matching database than 23andMe or AncestryDNA, which matters only if you are using them for genealogy rather than uploading to Haeckel.
FTDNA's autosomal chip test covers about 700,000 SNPs. Raw data downloads as a CSV from the myDNA section. Compatible with Haeckel. FTDNA also sells separate Y-DNA and mtDNA sequencing products (BigY-700, mtFull Sequence) that go much deeper into haplogroup resolution than a chip can — those produce BAM files rather than genotype tables and are not yet compatible with the Haeckel upload path, but Haeckel can still use any haplogroup calls you already have from those tests by looking them up in the ancestral databases.
A UK-based company whose chip covers roughly 640,000 SNPs with notably dense coverage of British Isles subregions. Raw data downloads as a .txt file similar in structure to 23andMe's format. Good choice if you have heavy British, Irish, or Northern European ancestry and want the most granular regional breakdown. The sub-regional granularity is a LivingDNA proprietary product — Haeckel runs its own ancestry inference from the raw markers regardless.
Type 2 — Whole genome sequencing (WGS)
Whole genome sequencing reads every position across your full 3.2 billion base pairs rather than a fixed panel. At 30× depth — the standard most consumer WGS providers use — each position is read on average 30 times from independently prepared DNA fragments, giving very high confidence in the calls. The result is a VCF (Variant Call Format) file listing every position where you differ from the reference genome, typically 4–6 million variants compared to the 700,000 on a chip. The WGS files are also larger: a compressed VCF from a 30× run is typically 1–4 GB.
The most widely used consumer WGS service. 30× Illumina short-read sequencing (150 bp paired-end) on GRCh38. Delivers FASTQ (raw reads), BAM (aligned reads), and VCF (variant calls). Upload the VCF to Haeckel for the fastest analysis — it is the most directly usable format. FASTQ also works but requires the pipeline to align and call variants internally, adding processing time. Nebula delivers files via their dashboard download section; files are large (VCF ~1–4 GB, FASTQ pairs ~60–100 GB combined) so use a wired connection or their download manager. Pricing has varied over time; check their current pricing page before ordering.
An Italian WGS company with strong European data-protection practices (GDPR-native). Offers 30× Illumina WGS with FASTQ + BAM + VCF delivery. Their Whole Genome 30x product (roughly €199 at time of writing) is the most relevant. Files are delivered via the MyDante dashboard. A slower turnaround than Nebula (often 6–10 weeks) and delivery communication can be patchy, but raw data quality is equivalent. A good option if you prefer a European data processor.
An aggregator platform that works with multiple accredited sequencing labs rather than running its own facility. They broker 30× WGS and deliver VCF + BAM files through their platform. Also allows you to upload existing chip or WGS files and run them through their own apps — independent of Haeckel. Compatible with Haeckel's VCF upload path. The main advantage is their large app ecosystem if you want third-party analysis tools in parallel.
Whole genome sequencing ordered through a physician or genetic counsellor at a CLIA-certified lab (Quest Diagnostics, LabCorp, Ambry Genetics, GeneDx, Genome Medical). The clinical VCF is the highest-quality file you can get — every call has passed clinical-grade quality filters and the variant interpretation is done against medically curated databases. If you have a specific health concern driving the test, this is the right route. Many insurance plans now cover WGS for medically justified indications (cancer, rare disease, family history of monogenic conditions). Ask the ordering physician to provide the raw VCF alongside the clinical report — the report alone is not uploadable. Not all clinical labs give patients access to their raw VCF by default; you may need to request it explicitly.
Type 3 — Long-read sequencing (Oxford Nanopore, PacBio)
Long-read sequencing technologies generate individual reads of 10,000 base pairs or more — compared to the 150 bp reads from Illumina short-read WGS — which fundamentally changes what the data can reveal. Long reads can span repetitive regions that short reads cannot, detect structural variants (large insertions, deletions, inversions, copy-number changes) that short reads miss, fully phase your two haplotypes (which copy came from which parent), and, in the case of Oxford Nanopore, directly detect DNA methylation without a separate bisulfite conversion step.
ONT makes the MinION (a portable USB sequencer, ~$1,000 device cost plus consumables), the GridION, and the PromethION (large-scale instrument used in clinical and population genomics). The Q20+ chemistry released in 2022 achieves modal read accuracy above 99%, comparable to Illumina for most variant types. The reads average 15–50 kb with some exceeding 1 Mb (ultra-long reads used for de novo assembly). ONT data comes as FASTQ files (or raw pod5 signal files before basecalling). The platform is not available direct-to-consumer: you need access to a university, hospital, or research facility that runs a PromethION or MinION. Some research programs make participant sequencing data available for personal use — ask if you are enrolled in any genomics cohort study. Haeckel will support ONT FASTQ once the long-read variant calling pipeline is validated.
Pacific Biosciences' HiFi chemistry produces reads of 15–20 kb with per-read accuracy of 99.9% — the highest raw accuracy of any long-read platform. The Revio system generates roughly 90 Gb of HiFi data per day, enough for 30× WGS on a human genome. The data is delivered as FASTQ or BAM (CCS consensus reads). Like ONT, PacBio is not consumer-accessible and requires institutional infrastructure. Some commercial services (e.g. ColoSSe Genomics, Hudson Alpha Institute) offer PacBio sequencing to academic or clinical customers. Coverage pricing is typically $1,500–4,000 for a 30× human WGS run. Future Haeckel support is planned.
What each test type unlocks in Haeckel
- Chip array (23andMe / AncestryDNA / MyHeritage / FTDNA / LivingDNA): ancestry inference with 7 super-populations and 62 sub-populations; haplogroup assignment (Y-DNA and mtDNA); Neanderthal and Denisovan archaic ancestry; all 15 trait predictions; pharmacogenomics across 12 genes; nutrigenomics across 10 dietary categories; all 9 polygenic risk scores; health variants from ClinVar; APOE isoform; runs of homozygosity (consanguinity and endogamy estimation). Coverage: roughly 700,000 positions, all common variants, ~97% of Haeckel features.
- 30× WGS VCF (Nebula / Dante Labs / clinical): everything above, plus rare variant discovery across the full genome; better pharmacogenomic star-allele coverage in gene regions not well-tagged by chips (e.g. CYP2D6 structural duplications); more variants contributing to each polygenic score (up to 5–6 million vs 700,000 on a chip); higher ancestry resolution in admixed populations where rare chip markers create noise. Coverage: ~5–6 million variant positions. All Haeckel features plus rare-variant health analysis.
- Long-read WGS (ONT / PacBio — future): everything above, plus repeat expansion detection, full diploid phasing, methylation profiling. Features pending pipeline development.
Decision guide
- Under $100 budget: AncestryDNA ($99) or MyHeritage ($79). Both deliver the full chip dataset Haeckel needs.
- $100–200 budget: 23andMe Health + Ancestry ($199). Their own health reports plus Haeckel analysis.
- $200–350 budget: Nebula Genomics 30× WGS (~$299). Biggest jump in data quality. Most complete Haeckel output.
- Already tested + want more: add a WGS test (Nebula or Dante). Upload the WGS VCF and it replaces the chip as your active analysis.
- Primary goal is family matching / genealogy: AncestryDNA has the largest database. LivingDNA if you are heavily British Isles.
- Primary goal is Y-DNA haplogroup depth: FTDNA BigY-700 (separate $399 Y-only test). Not yet uploadable to Haeckel but useful for haplogroup confirmation independent of Haeckel.
- Medical or clinical focus: consult a genetic counsellor first, order through a physician, request the raw VCF.
How to download your raw data from each provider
23andMe
- Log in at 23andme.com.
- Click your name in the top-right corner → Settings.
- Scroll to the "23andMe Data" section.
- Click "View" → "Browse Raw Data" → "Download Raw Data".
- Choose "All DNA" and confirm with your password.
- A zip archive is prepared. Most accounts receive an immediate download link; some receive an email link within 24 hours.
- Extract the zip. The file you want is named genome_[YourName]_v5_Full_[date].txt (or similar).
- Upload this .txt file directly to Haeckel. Do not modify it.
AncestryDNA
- Log in at ancestry.com.
- Click "DNA" in the top navigation → "Your DNA Results Summary".
- Click "Settings" (or the gear icon next to your kit name).
- Scroll to "Download DNA Data" → click Download.
- Confirm via the two-step email verification Ancestry sends to your registered address.
- A zip archive downloads. Extract to find a .txt file with columns rsid, chromosome, position, allele1, allele2.
- Upload this file to Haeckel.
MyHeritage DNA
- Log in at myheritage.com.
- Click "DNA" in the top menu → "Manage DNA kits".
- Click the three-dot menu ("...") next to your kit → "Download raw DNA data".
- An email with a download link is sent to your registered address (usually within minutes).
- Download and extract the zip. Upload the CSV file inside.
FamilyTreeDNA (Family Finder)
- Log in at familytreedna.com.
- Click "myDNA" in the navigation.
- Click "Download Raw Data" → select "Genome" (the autosomal data, not Y or mtDNA).
- A zip file is prepared and a download link is emailed. Extract and upload the CSV.
LivingDNA
- Log in at livingdna.com.
- Go to your Dashboard → click "Download" under your kit.
- Choose the raw data download option (not the ancestry report PDF).
- Extract and upload the .txt file.
Nebula Genomics (WGS)
- Log in at nebula.org.
- Click "Explore" → "My Files" in your dashboard.
- You will see FASTQ (raw reads), BAM (aligned reads), and VCF (variant calls) listed.
- Download the VCF file for the fastest Haeckel upload. It is typically 1–4 GB compressed (.vcf.gz).
- Upload the .vcf.gz directly — Haeckel detects and decompresses gzip files automatically.
- If you want to upload FASTQ for the most complete analysis (slower processing), download the paired FASTQ files instead.
Dante Labs (WGS)
- Log in at dantelabs.com → click "MyDante".
- Find your order and click "Download Files".
- Download the VCF file (or FASTQ pair if you want raw reads). Dante delivers via a download portal with a time-limited link.
- Use a download manager for large files — a 4 GB VCF on a slow connection benefits from a tool that can resume interrupted downloads before the link expires.
- Upload the VCF to Haeckel.
Prices change — check before ordering
Consumer genomics pricing is competitive and changes frequently. 23andMe in particular has changed its pricing multiple times and has offered the Ancestry Service for as low as $49 during promotions. AncestryDNA routinely discounts to $59 for the DNA-only kit during Black Friday and Amazon Prime Day sales. Check each provider's website for current pricing before ordering. The prices listed in this article reflect typical pricing as of mid-2026 and should be treated as approximate guidance rather than a precise quote.
- 23andMe raw data export documentation: customercare.23andme.com
- AncestryDNA raw data documentation: support.ancestry.com
- Nebula Genomics whole genome sequencing: nebula.org/whole-genome-sequencing
- Dante Labs whole genome: dantelabs.com/products/whole-genome-sequencing
- Oxford Nanopore Technologies — Q20+ chemistry: nanoporetech.com/platform
- Pacific Biosciences HiFi sequencing: pacb.com/smrt-science/smrt-sequencing
- CLIA-certified clinical lab ordering: genome.gov/about-genomics/policy-issues/Direct-to-Consumer-Genetic-Testing
I am thinking about getting a DNA test. Based on what I want to know about myself, what type of test would give me the most useful results in Haeckel?