Promethease is gone. Here are the real alternatives for interpreting your DNA raw data
For about fifteen years, Promethease was the answer to "I have my 23andMe raw data, now what?" It was cheap (often $12), thorough (it pulled from SNPedia, which itself drew from PubMed), and stripped of the marketing language that consumer genetic-health products tend to wrap around their reports. It was beloved by the technically curious for exactly those reasons.
Promethease is no longer accepting new uploads. SNPedia, the curated database it was built on top of, has had limited maintenance for a while. The result: anyone with a 23andMe, AncestryDNA, or MyHeritage raw file is now looking for somewhere else to take it. If you landed here searching for a Promethease alternative 2026, or more broadly a SNPedia alternative for third-party DNA analysis, this is the honest map of the landscape.
This post compares the realistic alternatives, honestly, including ours. The TL;DR: there are four serious options, they're each good at different things, and you should pick based on what you actually want.
What Promethease was actually good at
Worth being specific about. Promethease was strong on three axes:
- Comprehensive variant coverage. It surfaced thousands of variants per upload, not the small curated list 23andMe shows.
- Citation-anchored. Every claim linked back to a SNPedia entry, which itself sourced PubMed studies.
- Editorial restraint. It didn't tell you what to do. It told you what the literature said.
Any honest alternative has to be evaluated against those three. Most current options succeed on one or two and fail on the third.
The current landscape
Expressive (us, disclaim accordingly)
What it is: A privacy-first DNA analysis platform built around raw DNA file interpretation. You upload your raw file; we map every variant to the published research and produce a structured report with per-variant evidence quality, citations, and recommendations only where the literature supports them. We don't prescribe, we describe.
What it's good at: Citation density. Every claim anchors to a specific PMID on PubMed, a PharmGKB clinical annotation, an NHGRI-EBI GWAS Catalog row, or a ClinVar interpretation. Privacy posture is the most defensible in this category, we use AES-256-GCM encryption with keys derived from your wallet signature, and we cannot decrypt your data server-side without your participation. Your genome stays yours. We're explicit when the evidence is thin and refuse to invent recommendations.
What it's not good at: Speed of insight. We surface the studies, the effect sizes, and the hedging. If you want "take this supplement," we're the wrong service. If you want "what does the research actually say about my variants and how confident is it," we are exactly the right service.
Pricing: In private beta during 2026, expected freemium with paid tiers for full report access at public launch.
Genetic Lifehacks
What it is: A blog plus paid subscription with a raw-data report layered in. Founded by Debbie Moon, with a strong reputation in the biohacker community for thorough writeups on individual gene variants.
What it's good at: Depth on specific genes. The writeups are long, well-researched, and grounded in actual studies. The community lean is biohacker-friendly without crossing into supplement-grift territory.
What it's not good at: Programmatic coverage of every variant in your file. The reports tend to focus on the most well-studied variants, not the long tail. Citations are present but not as prominently anchored per-claim as Promethease was.
Pricing: Subscription, around $10/month at last check.
SelfDecode
What it is: A larger commercial product that takes raw data and produces health, fitness, mood, and longevity reports along with personalized recommendations.
What it's good at: Production polish. The reports are visually clean, the UX is smooth, and the breadth of trait coverage is significant.
What it's not good at: Editorial restraint. SelfDecode tends to surface recommendations as confidently as the underlying evidence allows, which sometimes means recommending things on thin evidence. The privacy policy is also looser than the privacy-conscious users we hear from would prefer, they retain data in ways that, while not unusual for the industry, are notably less restrictive than the model Expressive uses.
Pricing: Recurring subscription, varies by tier.
Genomelink
What it is: A platform that takes raw data and offers a catalog of individual trait reports you can purchase a la carte or via subscription.
What it's good at: Variety. Hundreds of small reports across health, fitness, personality, and ancestry. Good for people who want to dabble in genetic interpretation without committing to a single deep report.
What it's not good at: Citation density. The reports tend to summarize the conclusion of a paper rather than expose the underlying evidence quality. Less suitable for the user who wants to see exactly which study a finding came from, or to click through to the underlying dbSNP record.
Pricing: Per-report and subscription tiers.
What about Genetic Genie?
It's still around for a smaller, focused use case, primarily MTHFR and a handful of related methylation-pathway variants. If those are the only variants you care about, it's free and quick. For everything else, look elsewhere.
Choosing among them
The honest framing:
- If you want everything Promethease offered, with stronger privacy and explicit evidence-quality flags: that's us.
- If you want long-form writeups on individual genes and you trust the biohacker community framing: Genetic Lifehacks.
- If you want a polished UX and you don't mind broader interpretive license: SelfDecode.
- If you want to buy individual trait reports a la carte: Genomelink.
- If you only care about MTHFR: Genetic Genie or any methylation-focused tool.
There is no "best" here. Each tool has a different center of gravity and the right one depends on what you're actually trying to get out of it.
What didn't make this list (and why)
There are several categories of products we deliberately don't recommend:
- "Personalized nutrigenomics" subscriptions that promise meal plans based on your DNA. The underlying nutrigenomic evidence base is thin enough that these products almost always overclaim. The recommendations would be similar regardless of your actual variants.
- "Personalized fitness" products with the same problem at a different layer. Genetics influences athletic response but the effect sizes are too small to drive specific training-plan changes for most people.
- "DNA-matched skincare" and similar consumer-product lines. Marketing wrappers on standard products.
These exist, they have marketing budgets, and they will appear in your Google results when you search for Promethease alternatives. The reason they're not in this comparison is that they're not actually solving the "I want to know what my raw data says" problem. They're solving the "I want to sell you a product" problem.
What you should do this week
Whichever tool you pick, two things are worth doing right now:
- Download your raw data file. Even if you're going to upload it to a service, save your own copy first. We wrote a walkthrough on the 23andMe raw export. Same idea for AncestryDNA and MyHeritage with minor menu differences.
- Read the privacy policy of the service you upload to. Not the marketing page. The actual policy. Look for what they say about data retention, research use, third-party sharing, and account deletion. The differences across products in this category are significant.
If you want to try Expressive as your SelfDecode alternative or Genomelink alternative, or simply as a way to interpret 23andMe raw data without surrendering your file to an ad-targeted profile, you can sign up here. We're in private beta in 2026 but the waitlist moves fast. The corpus we've been building covers ~74,000 SNPs and ~39,000 genes with citation-anchored summaries, and the personal report layer maps your raw file against it. Evidence quality always visible.
Want updates when we ship new variant pages or a research deep-dive? Read the latest issue or get notified about early access.