filmov
tv
What they didn't tell you about NEANDERTHAL admixture

Показать описание
Everyone who has access to the internet and has an interest in genetics knows that Eurasians have denisovan and neanderthal admixture.
This is something we all pretty much know, and take at face value, without ever doubting. Experts say so, therefore it must be true!
What we often don’t realize is that experts own opinions change over time, one decade they say one thing - another decade when their technology changes they say something else. Another thing we don’t realize is that experts are humans just like us. They have biases, they have their own agendas, they make mistakes.
At the end of the day, just because a Harvard study concluded something, does not mean that this conclusion can not be scrutinized if it has obvious flaws.
I see some very obvious logical flaws with statements about neanderthal or denisovan admixture in eurasians, or any archaic admixtures in any modern human population for that matter. In this video I will discuss them.
Admixtools and QPADM can supposedly detect neanderthal admixture in Europeans and East Asians, but I am highly skeptical of the validity of such models. These trends could and likely do simply reflect genetic drift in Eurasians that isn’t found in Sub Saharan Africans. QPADM models can be manipulated to display just about anything and need to be taken with a grain of salt.
Additionally, having worked with neanderthal DNA files in the past, I noticed a very peculiar trend that neanderthals resemble Sub Saharan Africans, not Eurasians. On every GEDmatch calculator and PCA based calculator they score Sub Saharan Africans. On my own ethnicity calculator they score identical to Sub Saharan Africans.
Since we do not have a Eurasian sample without this supposed neanderthal admixture to compare the rest of eurasians to, all these QPADM models showing neanderthal and denisovan admixture become rather irrelevant - all they do is show that Eurasians have diverged from Sub Saharan Africans, not that we have neanderthal or denisovan admixture.
The most robust way to establish that some sort of admixture happened is by DNA segment size. That’s how 23andMe establishes the ancestry timeline, or when your ancestors from a certain ethnicity became a part of your family tree. Since DNA is inherited in chunks, the larger the chunks the more recent this admixture is. You could be 25% french for example, but if those 25% are all from one granddad - then you would have only a couple very large chunks of french DNA in your genome. Whereas if your 25% french ancestry was from a bunch of ancestors from 10 centuries ago who mixed a long time ago, your french chunks would be small and more numerous.
Here’s what I’m getting at though. The chunks of neanderthal admixture we would have after 40 thousand years of mixing would be so small that it is impossible to establish whether it is the result of admixture or genetic drift.
The final nail in the coffin is the shortage of high quality neanderthal DNA samples. We can not build an allele frequency database for them like we can for modern populations, we can not study them the way we can study modern human groups.
What this means in the context of the Q.P.A.D.M. models mentioned earlier, is that an affinity to these low quality neanderthal samples such as the female skeleton from vindija does not imply an affinity to neanderthals as a whole, but rather implies a deviation from Sub Saharan Africans used as outgroup in the Q.P.A.D.M. model
This is something we all pretty much know, and take at face value, without ever doubting. Experts say so, therefore it must be true!
What we often don’t realize is that experts own opinions change over time, one decade they say one thing - another decade when their technology changes they say something else. Another thing we don’t realize is that experts are humans just like us. They have biases, they have their own agendas, they make mistakes.
At the end of the day, just because a Harvard study concluded something, does not mean that this conclusion can not be scrutinized if it has obvious flaws.
I see some very obvious logical flaws with statements about neanderthal or denisovan admixture in eurasians, or any archaic admixtures in any modern human population for that matter. In this video I will discuss them.
Admixtools and QPADM can supposedly detect neanderthal admixture in Europeans and East Asians, but I am highly skeptical of the validity of such models. These trends could and likely do simply reflect genetic drift in Eurasians that isn’t found in Sub Saharan Africans. QPADM models can be manipulated to display just about anything and need to be taken with a grain of salt.
Additionally, having worked with neanderthal DNA files in the past, I noticed a very peculiar trend that neanderthals resemble Sub Saharan Africans, not Eurasians. On every GEDmatch calculator and PCA based calculator they score Sub Saharan Africans. On my own ethnicity calculator they score identical to Sub Saharan Africans.
Since we do not have a Eurasian sample without this supposed neanderthal admixture to compare the rest of eurasians to, all these QPADM models showing neanderthal and denisovan admixture become rather irrelevant - all they do is show that Eurasians have diverged from Sub Saharan Africans, not that we have neanderthal or denisovan admixture.
The most robust way to establish that some sort of admixture happened is by DNA segment size. That’s how 23andMe establishes the ancestry timeline, or when your ancestors from a certain ethnicity became a part of your family tree. Since DNA is inherited in chunks, the larger the chunks the more recent this admixture is. You could be 25% french for example, but if those 25% are all from one granddad - then you would have only a couple very large chunks of french DNA in your genome. Whereas if your 25% french ancestry was from a bunch of ancestors from 10 centuries ago who mixed a long time ago, your french chunks would be small and more numerous.
Here’s what I’m getting at though. The chunks of neanderthal admixture we would have after 40 thousand years of mixing would be so small that it is impossible to establish whether it is the result of admixture or genetic drift.
The final nail in the coffin is the shortage of high quality neanderthal DNA samples. We can not build an allele frequency database for them like we can for modern populations, we can not study them the way we can study modern human groups.
What this means in the context of the Q.P.A.D.M. models mentioned earlier, is that an affinity to these low quality neanderthal samples such as the female skeleton from vindija does not imply an affinity to neanderthals as a whole, but rather implies a deviation from Sub Saharan Africans used as outgroup in the Q.P.A.D.M. model
Комментарии