Neanderthal interbreeding
So the first draft of the Neanderthal genome has now been published (http://www.sciencemag.org/cgi/content/full/328/5979/710). The authors found that Neanderthals contributed 1-4% to the genomes of non-Africans. They use SNPs and five full genomes from different ethnic groups to arrive at this conclusion. Aspie-quiz has also found that all non-Africans have similar levels of Aspie-traits, while people of African descent have much lower levels of Aspie-traits (1/6th for African-Americans), and might even totally lack these traits when unmixed with non-Africans.
The study claims that admixture most likely occured in Middle East, but this is not supported by Aspie-quiz. Aspie-quiz instead seems to indicate that East Asians mixed with Neanderthal before Europeans. Aspie-quiz also give native American Indians the highest scores of any population, and thus it seems like they have the largest Aspie rate in the world. South Asians are lower than Europeans and East Asians. There is a cline in Eurasia where the north has much higher Aspie-rates than the south. This could be compatible with interbreeding in the Middle East, and later African migration into south Asia, but it matches the “southern migration route” of modern humans far better. Under this scenario, Africans and south Asians are largely devoid of primary interbreeding with Neanderthal. Interbreeding started in Central Asia, and this population then split into a western part that ended up in Europe and formed Caucasians, and an eastern part that ended up in East Asia and formed Mongoloids and Native Americans. Native Americans then got isolated from further gene-flow from South Asia, and thus today has the world’s highest level of Neanderthal ancestry.
The 1-4% of Neanderthal ancestry probably refers to neurotypicals only. Typical Aspie scores for neurotypical people are around 40-50. Typical Aspie scores for Aspies are three times larger. We would therefore assume that Aspies perhaps has 5-15% of Neanderthal ancestry using this as an indication. OTOH, if we assume that 1-4% is the average Neanderthal ancestry in Eurasia, and that all of this is attributable to Aspie traits, and the proportion of Aspie traits is around 10%, it would mean that Aspie traits have 10-40% of Neanderthal ancestry.
Another recently published article on the same theme came to similar conclusions (http://www.nature.com/news/2010/100420/full/news.2010.194.html). Although, here the authors see two different hybridization events. This scenario in fact is supported in Aspie-quiz. Not only is there a signature of Neanderthal interbreeding leading to the introduction of the Aspie-traits, but there is also a signature of the introduction of “neurotypical compulsion” traits that essentially describes modern humans.
To resolve these issues there is a need to sequence full genomes of Aspies.
May 12th, 2010 08:25
There’s been some work in figuring out the Aspie genotype.
The actual paper is…
“Genes related to sex-steroids, neural growth and social-emotional behaviour are associated with autistic traits, empathy and Asperger Syndrome”
http://people.pwf.cam.ac.uk/bc249/papers/chakrabarti_2009_autismres_GeneAutismSexSteroidNeuralGrowthSocial.pdf
ScienceDaily also did a write up of it here…
http://www.sciencedaily.com/releases/2009/07/090715101427.htm
The sequenced Neanderthal genome is online, so you could take a look to see if it matches any of the alleles suspected to be part of the Aspie genotype.
May 16th, 2010 14:53
@Leif: This is OT. I was wondering if you can make your Aspie-Quiz individual-level data available (if not publicly, to me personally.)
That is, one row of data per person, with autistic status (non-autistic, diagnosed, self-diagnosed, etc.) plus answers to each question. I don’t even need to know what the questions are so long as they are identifiable later (#1, #2, etc.)
Also, what method do you currently use to assign a weight to each question?
May 17th, 2010 05:26
Yes, I can give interested people access to raw-data for Aspie-quiz. I have done this in the past to people that I feel have enough knowledge to analyse the material properly.
A complication is that there is not one data-set for Aspie-quz, but currently over 50. However, the most reasonable data to analyse is final version II, which (currently) has a little over 22,000 answers.
Weights to questions are assigned based on average factor loadings in previous versions. This is described in the (currently unpublished) paper about Aspie-quiz, which I also give out to interested people that promise to keep it confidential. You really need to read this paper before working on the raw-data.
May 17th, 2010 06:01
@Leif: I’m assuming the 50 data sets correspond to different versions of the test? I’d probably be interested in either the latest version or the version with most data. What would you recommend?
These days I’m interested in certain machine learning problems, e.g. classification problems (like spam filtering.) You probably know what I’m talking about. There are several techniques that can be used to approach these problems, like Bayesian techniques, logistic regression, etc.
Scoring screening tools is a lot like a classification problem.
If I find something of interest in my analysis of your data, I will of course post that on my blog.
May 17th, 2010 06:46
Yes, the 50 different data sets corresponds to different versions. You probably want the main data set from final version II, as I stated above. It isn’t the largest data set (the international version with 54,000 answers is, but these answers have been given in different languages, and so might not be suitable for analysis).
I have prepared a xls-document for you (it is 23 MB), and a compressed xls-document (2.7 MB). Give me an email-address that can handle 2.7 MB attachments and I’ll send you the data.
May 17th, 2010 07:32
@Leif: You can send it to the email address I provide with the comment. I understand it can handle up to 10Mb attachments. Thanks.
May 19th, 2010 05:05
You may already know the Topology theory:
http://www.autismtheory.org/topotheory.html
The theory states, that that people living far away from the “core” of the society evolve autistic traits, because these traits are more important for their survival than evolving in social traits.
People living in low population density areas build traits that serve their own survival, while people living in densely populated areas (like in towns) require more social skills.
The highest score of the native american indians matches into this theory well. Considering that most of their history means living in low population density, and later, after europeans arrived, amerindians kept living with low rate of breeding with caucasians it is not surprising to find such a high value in their population.
Both theory suggest that adopting to a different environment (with low pop. density) results in emerging concentration of autistic or aspie traits.
Maybe the high value of the american indians in this quiz is more of a result of being isolated from the dense population.
May 19th, 2010 05:25
Yes, I certainly know about the Topology theory of autism. The predictions are similar to the Neanderthal theory because Neanderthals also lived in sparse populations. What differs is that the Topology theory propose that these traits evolved in our species, while the Neanderthal theory propose the traits were introgressed from another species. We have to look at other things to decide which of the theories have more merit. Personally, I don’t think the Topology theory is compatible with the differences in nonverbal communication that aspies shows and which are central to the autistic spectrum.
June 10th, 2010 15:58
I’m more than very happy of that news!
Our noble Grand-Father, Homo S. Neanderthalensis still alive in us. And the felling I had when I saw is rebuild face for the firts time can only make me more confident that aspie are the modern sons of this noble man. It’s what I feel .
Hag
July 27th, 2010 00:21
Lots of us online read the Science study as saying Autistic and certain other genes came from Neanderthals. But it’s mostly being reported to say that those specific genes being mentioned, are theorized to have actually developed after and (seemingly) independent of Neanderthal interbreeding. I’m a layman but a very-well-read one (probably Aspie, for what it’s worth!). Can anyone address what seem to be opposing readings of this part of the report? Thanks very much, Pete.