Comments on: Ancestry.com's AncestryDNA won't give you your raw data

By: Gale Gorman

Gale Gorman — Wed, 06 Feb 2013 15:18:00 +0000

I just got my so-called results this morning and still can’t believe they charged money to tell me information found on my public tree that I posted. What a scam!

By: Hannah Grant

Hannah Grant — Wed, 06 Feb 2013 09:35:00 +0000

I agree. It feels as if I have wasted my hundred dollars on thier product. I believe the most disappointed consumers will be the average white female with little genetic variety. That is to say the variety wouldn’t be visible except in the raw data. Then again Ancestry still claims that thy don’t have a polished product even though it’s gone live. The only thing I can look forward to is thier promise of future updates.

By: Paul Wright

Paul Wright — Tue, 18 Sep 2012 22:01:19 +0000

No raw data and no detail on segment shares or methodology/thresholds used to determine match leave much to be desired. Until corrected will be 3rd out of the 3 companies offering test that I would recommend anyone to purchase.

By: pconroy

pconroy — Tue, 18 Sep 2012 18:45:31 +0000

@10 Doug,

We certainly know that 23andme has sequenced many more people than ancestry.com, though I don’t know the actual ratio

Llast I saw AncestryDNA was selling like hot cakes, and had an estimated 50,000 users in < 6 months. It took 23andMe and FTDNA years to reach that mark. I’d estimate 23andMe probably has between 150,000 – 200,000 participants, which would be 3 or 4 times the Ancestry DNA database only.

Of course to answer my own question – I guess it's possible that 23andMe is "capping" my number of reported matches – but then it would have to be by a lot?

By: Douglas Knight

Douglas Knight — Tue, 18 Sep 2012 17:56:31 +0000

If 23andme has tested 15x as many people as ancestry.com, they should find 15x as many 4th cousins, 5th cousins, etc. We certainly know that 23andme has sequenced many more people than ancestry.com, though I don’t know the actual ratio.

By: pconroy

pconroy — Tue, 18 Sep 2012 17:21:53 +0000

Of course I should say that I too will demand my Raw Data be made available from AncestryDNA, if it is not done in the next few months.

By: pconroy

pconroy — Tue, 18 Sep 2012 17:19:03 +0000

@6 Doug,

That would account for the 50% MORE number of total “Relatives”, but what I’m really interested in is the 1,533% LESS close relatives identified – how can this be?!

By: Douglas Knight

Douglas Knight — Tue, 18 Sep 2012 17:18:32 +0000

(I don’t know the criteria either company uses for identifying relatives. I got the numbers 7th & 10th by assuming that you have twice as many nth cousins as n-1-th cousins. And I ignore everyone not on the same generation. The point is that different thresholds could easily produce this situation.)

By: Douglas Knight

Douglas Knight — Tue, 18 Sep 2012 17:15:31 +0000

pconroy, it’s probably just that 23andme only counts 7th cousins as relatives, while ancestry.com counts 10th cousins.

Another possibility stems from the fact that Ancestry.com’s main product is not DNA testing, but software for making family trees. They may be counting as distant relatives people in the trees of those who have been tested.

By: Razib Khan

Razib Khan — Tue, 18 Sep 2012 17:13:25 +0000

#4, have you ever downloaded your raw genotype file from 23andme? it’s easy to convert it into other formats (pedigree format) and do further analytics. though it isn’t totally raw, as they get it in a different format from illumina. i had no idea that only sequence data is termed ‘raw.’ is this for real?

By: Patrick

Patrick — Tue, 18 Sep 2012 16:53:18 +0000

What do you mean by raw data?

If it is anything like 23andMe, the reason that you can’t get a copy of the data is because they don’t actually sequence your genome. They just use a battery of tests to identify particular SNPs.

By: pconroy

pconroy — Mon, 17 Sep 2012 21:44:27 +0000

Yes, this product is currently Beta.

I am estimated as 95% British and 6% Unknown – I sent them an email asking what were the possibilities for Unknown a few weeks ago and got no response.

Another thing to ponder is that on 23andMe I have something like 1,100 Relatives, of which an estimated 92 are 4th cousins or less.

Whereas on AncestryDNA, I have something like 1,450 Relatives, of which an estimated 6 are 4th cousins or less – at > 95% confidence.

So AncestryDNA has about 50% MORE Relatives identified, but has 1,533% LESS close relatives identified?!

I’m not sure how to evaluate this, on the one hand it may be that Ancestry are overly cautious, or 23andMe are overly optimistic, while on the other hand it may be in part due to the databases themselves, where many of 23andMe’s customers – at least originally – came from a medical/technical or professional background, like my own, whereas many of AncestryDNA customers were existing Ancestry customers and a more average American population?!

Any ideas anyone?

By: April Brown

April Brown — Mon, 17 Sep 2012 08:20:37 +0000

Let’s hope it’s an artifact of being in beta – I’ve been on the back side of enough software development projects to smell the corporate-speak apology that results from conflicts among marketing and dev and test. Bloggers like you complaining and publicizing the shortcoming is exactly the right way to get them to reshuffle their ‘list of priorities’. I’d bet there’s something pointless like “customizable colour palettes for charts” up higher on their list than “provide raw data downloads.”

By: Taneya

Taneya — Mon, 17 Sep 2012 05:19:35 +0000

I really wish they would give the raw data and until they do, I probably will not buy their testing service. Thanks for the post.