Provisional Spring 2026 issueScores are based on public information, pending reader-supported independent testing.Methodology →
genomereviews
Volume IX · Issue 4 · Spring 2026Independent · Reader-supported · No affiliate linksSaturday, April 18, 2026
How we review

Our methodology, written down so you can argue with it.

We take no affiliate commissions, we buy every kit with reader-supported funds, and we publish the spreadsheet. Below: what we score, how we weight it, and what it takes for a company to land in each tier.

Every issue we review the consumer whole genome sequencing market from scratch. No standing favorites, no grandfathered clauses. A company on last issue's Editor's Choice can fall to Proceed With Care by the next if our purchased-with-reader-funds kit comes back with a 22× average where 30× was promised.

The five sub-scores

How the 100-point composite is built.

DimensionWeightWhat we measure
Accuracy20%Agreement with a reference sample sequenced at 150× on a second platform; variant-level F1 against the GIAB truth set.
Coverage20%Mean depth delivered vs. promised; uniformity across chromosomes; percentage of callable bases at 20×.
Reports20%Clinical soundness, visual clarity, how well uncertain findings are contextualized, and whether a clinician could act on them.
Privacy20%Data residency, retention controls, who can re-identify you, what happens on acquisition, and whether deletion is real.
Value20%Price vs. delivered coverage, report depth, raw file access, and included clinician interaction.
The four tiers

What a badge actually means.

Editor's Choice
≥ 90

At least 90 on the composite, ≥ 85 on every sub-score, and no unresolved operational red flags in the last two issues.

Recommended
78–89

Strong across the board with clear caveats the reader should know about before purchase.

Niche Pick
68–85*

Excellent on at least one dimension that matters intensely to a specific buyer (e.g. Y-DNA genealogy) — but not a general recommendation.

Proceed With Care
≤ 75

The science is sound but reliability, support, or operational track record raises real questions. Buy eyes-open.

Our process, per company

Six purchases. Three reviewers. One spreadsheet.

  1. Six kits purchased across the quarter from three different addresses.
  2. Each reviewer activates independently and records time-to-each-milestone.
  3. Raw FASTQ/BAMFASTQRaw sequencing read file. The unprocessed output of the sequencer. files are downloaded where offered and compared to our reference.
  4. Reports are scored blind by a panel that includes a board-certified genetic counselor.
  5. We submit a data deletion request; 90 days later, we verify it was honored.
What we will never do
  • —  Take affiliate or referral commissions.
  • —  Accept free kits from companies we review.
  • —  Let a company preview a review before publication.
  • —  Accept equity, advisory positions, or honoraria from anyone we cover.
Corrections & disputes

We make mistakes. Here's how we fix them.

Corrections run at the top of every issue and are permanently appended to the affected review. Companies that dispute a finding can submit documentation; if we update a score, the revision is dated and the old score remains visible. Ghost-edits are disqualifying.

corrections@genomereviews.com
© 2026 Genome Reviews · ISSN 2769-4102 · Press T to toggle Journal ↔ Terminal