Wednesday 25 July 2012

23 And Me and Me: Part 3

In which I get a nice surprise.

23andMe advertise that it takes 2-3 weeks for results to become available, after sending your saliva sample. I must have caught them in a slow period, because my results have started to come through only a week after they received my spit tube.
Clearly, some analyses take more compute or result preparation time than others, as (most of) my ancestry results have come through, but the health-related results are still unavailable as I type. The complete set of raw results was available for download so, of course, being what is technically known as a nosey bleeder, I've already got those. (Sorry, Apple. More interesting than Mountain Lion).

I'm A African^H^H^H Neanderthal

The results that are available have already thrown up some interesting information. For instance, I have a higher incidence of SNPs potentially related to Neanderthal DNA than most people who've been sequenced by this company in Northern Europe. There's even a jaunty graphic for it:
I feel irrationally proud of this. Recent research suggests that Neanderthals were sensitive wee souls who had art and buried their dead, rather than the thuggish cavemen I was led to believe when I was younger. I can even feel some good old middle-class Homo sapiens guilt at having wiped them out, with a tinge of honour at secretly carrying on some of the hairy tool-shy line.
Not that this is likely to have a significant impact on me, other than having a slightly longer second toe than usual. As you can see...

Paternity Suits

My paternal line - as indicated the Y chromosome SNPs -  is apparently resolutely Western European: R1b1b2a1a. And, not entirely surprisingly given what I know of recent family history, this is consistent with being sedentarily British. More pretty graphics show the estimated spread of the larger haplotype group R1b1b2 around 500 years ago:
The subgroup I'm in looks quite messy, from the brief reading I've done, so I'm glad I've grabbed the raw data. I'll try to make sense of it later.

Does Your Mother Know?

My maternal line - as indicated by mitochondrial SNPs - is interesting. I wasn't expecting that it would be so rare in Western Europe:
This haplogroup appears to be most common in the Caucasus, Iran/Iraq, Israel, Lebanon, and Turkey, with an intriguing signal from Finland. I'll be very interested to look at this in more detail, but there's no indication from my family tree (which my Dad has managed to trace back some way) where this line connects. I'm looking forward to finding out more about that.

On the whole...

I'm European. Results are trickling onto the website and, because even the H8 haplogroup counts as Europe, there's not a shred of colour indicating African or Asian DNA in my 'Ancestry Painting':
I'm neither disappointed nor surprised at this result, and I'm feeling that odd excited sensation that I get when I look at a new set of bacterial genomes, or a protein sequence family, that I want to find out more.
This could be addictive. On to the raw data!


  1. Hello. Nice post. Well-written!

    I am wondering how it's possible to have the markers for H8 but have zero Asian ancestry? I have this configuration too.
    You say H8 counts as Europe...but if it's rare in Europe and common in Central Asia - how does it count as Europe? Didn't that H8 ancestor come from Central Asia (first the Near East, then expanded to Asia) - so I can't wrap my head around this.

    Thank you.

  2. Hi Nomaya,

    Thank you for the comment, and the compliment. I'm glad you liked the post.

    Honestly, I was probably mistaken when writing the post, as looking at the description of H8 on 23andMe now, it clearly says "Region: Near East, Central Asia". Sorry about that.

    I should clarify that I wasn't saying that H8 indicates European ancestry, but rather that (I thought that) 23andMe seemed to classify H8 as "Europe". If you think that's weird, then you should see the entry requirements for Eurovision ( which includes Azerbaijan for a European contest ;)

    The way I understand having H8, but no other indication of Asian ancestry, is just that there's an unbroken maternal line back to an H8 maternal ancestor, likely from that region (or perhaps more recently, to the apparent enclave in Finland). As you'll know, the mode of inheritance for mitchondria is different from chromosomal DNA. It's almost certain that almost all chromosomal DNA from any particular long-ago ancestor has been shuffled out, by now. But mitochondrial DNA doesn't get shuffled out - it's inherited as a unit, so barring individual mutations, it's preserved over long time periods.

    If we imagine that your or my last H8 maternal ancestor lived 200 generations ago (maybe 4,000 years ago) then, assuming random shuffling and no interbreeding - which is admittedly unlikely - we'd expect very little chromosomal DNA from that individual to survive intact - only about (1/2)^200, which is about 0.000000000000000000000000000000000000000000000000000000000001%. But their mitochondrial DNA would likely be almost unchanged as it was passed down to you or me. Of course, with 23andMe we're really talking about alleles and their frequencies in populations, rather than actual inheritance of DNA fragments from individuals, but it's illustrative of how the difference you mention can arise.

    I don't know where the 500 years estimate on 23andMe's ancestry painting comes from, but that would represent the situation from only about 15 generations ago, I think (still long enough that you wouldn't see much DNA from any particular ancestor who lived at that time). In that respect, the ancestry painting reflects recent heritage, and mitochondrial DNA more ancient heritage.