Out-of-India - From Theory to Truth: Part 2

The Strategic Issues & International Relations Forum is a venue to discuss issues pertaining to India's security environment, her strategic outlook on global affairs and as well as the effect of international relations in the Indian Subcontinent. We request members to kindly stay within the mandate of this forum and keep their exchanges of views, on a civilised level, however vehemently any disagreement may be felt. All feedback regarding forum usage may be sent to the moderators using the Feedback Form or by clicking the Report Post Icon in any objectionable post for proper action. Please note that the views expressed by the Members and Moderators on these discussion boards are that of the individuals only and do not reflect the official policy or view of the Bharat-Rakshak.com Website. Copyright Violation is strictly prohibited and may result in revocation of your posting rights - please read the FAQ for full details. Users must also abide by the Forum Guidelines at all times.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

I think this is the image Rudradev meant to include in his post

By Gringer - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.p ... d=23655974
Image
Prem Kumar
BRF Oldie
Posts: 4218
Joined: 31 Mar 2009 00:10

Re: Out-of-India - From Theory to Truth: Part 2

Post by Prem Kumar »

Arun: that's what I am thinking too. But just because a particular mutation was found in the aDNA sample from the Steppes, doesn't mean its "Steppe DNA". This mutation could have very well evolved elsewhere and traveled to the Steppes.

Who's to say that the ancestor the Steppe DNA will not be found in aDNA from IVC?

The logic used by geneticists is:

1) Steppe aDNA has some mutation
2) This mutation is found in Indians too
3) So, Steppe people must have invaded India

The fact that there is no aDNA from India is conveniently ignored. If Indian aDNA is found to be more ancient & also have Steppe DNA, the same points (1) and (2) can be used to argue for OIT!
Rudradev
BRF Oldie
Posts: 4226
Joined: 06 Apr 2003 12:31

Re: Out-of-India - From Theory to Truth: Part 2

Post by Rudradev »

Prem Kumar wrote:Excellent points, Rudradev! Why don't you counter Vagheesh on Twitter? Or you should blog about this kind of buffoonery.

Had a couple of simple poochs:

1) Where did this "a large number of modern day Indians have Steppe DNA" come from? I am hearing it for the first time these days. Never before. And its being repeated as if its a well known folktale!
It is from this paper by Lazaridis et al, 2016. https://www.biorxiv.org/content/biorxiv ... 1.full.pdf
See the Extended Data Figure 4. They test a whole range of Indian population samples and see that there are markers shared between them and something they call "Steppe_EMBA".

What is "Steppe_EMBA"? Well, it's a statistical cluster of two things according to this paper.

The first thing is DNA that they extracted from 28 poor souls dated to the Early-Middle Bronze Age in different parts of the Eurasian Steppe. Whose ancestors could have come from anywhere, of course.

But the second thing is their "model" of the ancestry of this Steppe_EMBA population. As you can see in Extended Data Figure 5, it can be modeled as the mid-point of a line joining two positions on a "principal components analysis" graph. The two are (1) Iran Chalcolithic (i.e ancestors of the same Iranian Neolithic Agriculturists who are supposed to have contributed to "Indus Periphery" ancestry along with AASI!) and (2) European Hunter Gatherers.

These of course are ALSO statistical clusters. For all anyone knows their ancestors might all have come from India. But there is no evidence to say that, so we don't say it.

This whole business of "modeling" ancestry is interesting by the way. To borrow a metaphor from Shiv, consider a young man with (my) 14-inch pen1s and Angelina Jolie's smouldering eyes/full lips. He can be "modeled" as the love-child that Angelina Jolie and I had together during our torrid affair. This may or may not be true of course. (Though in this case, it is, God-promise.)
2) What is the meaning of "Steppe DNA"? Has it been proven that this particular mutation originated in the Steppes? Couldn't it also mean that this mutation could have originated elsewhere and moved to the Steppes (or) there might be a common parent from which this mutation originated. Words are important: if we call something as "Steppe DNA", then anyone else who has this DNA is automatically assumed to be a descendant of some Steppe-man
See above. Of course nothing like that has been proved. Absence of evidence is not evidence of absence... that is the key to rebutting any or all of this. The "Steppe_EMBA" samples could have been descended from chalcolithic farmers from Iran, descended in turn from neolithic agriculturists of Iran, descended in turn from a population in India. I agree with you that words are important (and in fact, even worse than "Steppe DNA" are terms like "Ancient Ancestral South-Indian HUNTER GATHERER" vs. "Neolithic Iranian AGRICULTURIST"). These terms are all consistent with a worldview in which ALL civilization (farming, tool-making etc.) came from the Levant and flowed east to influence more primitive peoples.

Which is exactly the worldview of the Bible... as a history-centric system, it is critically important for the Judeo-Christians to establish themselves as the source of history itself (they made tools and planted crops FIRST, everything else before them was "prehistoric").

Now if only our own guys would take a little time between chai-breaks to write up and publish the Rakhigarhi aDNA evidence, we would have something more than "absence of evidence" to argue from. Will it happen in our lifetime? Who knows. The Vagheesh et al paper is already a pre-emptive shot across the bow, intended to steer the "peer review" community towards institutional hostility against any conclusions the Rakhigarhi papers might in future espouse. The more time we waste, the further we fall behind.

And by the way, I did encounter Vagheesh on Twitter. He replied that I didn't know what I was talking about, and should publish my own challenge to their paper and have THAT peer-reviewed by David Reich's chamchas in the peer-review community.
Skanda
BRFite
Posts: 327
Joined: 18 Sep 2007 02:19

Re: Out-of-India - From Theory to Truth: Part 2

Post by Skanda »

Rudradev
BRF Oldie
Posts: 4226
Joined: 06 Apr 2003 12:31

Re: Out-of-India - From Theory to Truth: Part 2

Post by Rudradev »

A_Gupta wrote:
I had thought that they used the 4 groups - Swat SGPT+ Early Historic,Punjabi, Mala - as a quick way of identifying ancestral populations; and then used the resulting model on some (246 - 106) = 140 Indian groups and found a good match. It doesn't matterhow they arrived at their model as long as the resulting model fits well the groups with varying ANI/ASI proportions. Am I mistaken?

Thanks!
-Arun
No, you are right. The ancestral populations were defined using this skewed sample of 4 groups. Then, as Vagheesh said in one of his replies to me, they tested the model against the "entire Indian Cline!!!!" Meaning, 140 Indian groups after throwing out 106 because of "statistical noise" :mrgreen:

Is it surprising that their model got match hits with even 140 out of 246 Indian population samples?

Not at all. See, their model was Steppe_MLBA + Iran_N + AASI.

Steppe MLBA= 74% Steppe_EMBA and other things.

Steppe_EMBA = Iranian Chalcolithic and European Hunter-Gatherer.

Iranian Chalcolithic presumably descended at least in part from Iranian Neolithic (i.e. Iran_N, or Iranian agriculturists).

Iranian Agriculturists per the SAME paper share ancestry with ALL Indians, ASI, ANI, and the contrived "Indus Periphery".

So I'm actually more curious about the 106 samples that did NOT yield a match.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

Rudradev wrote: But the second thing is their "model" of the ancestry of this Steppe_EMBA population. As you can see in Extended Data Figure 5, it can be modeled as the mid-point of a line joining two positions on a "principal components analysis" graph. The two are (1) Iran Chalcolithic (i.e ancestors of the same Iranian Neolithic Agriculturists who are supposed to have contributed to "Indus Periphery" ancestry along with AASI!) and (2) European Hunter Gatherers.
For ready reference:
Image

Image
Last edited by A_Gupta on 10 Apr 2018 01:25, edited 1 time in total.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

From Figure 4 of the same 2016 Lazaridis et. al. paper, here's where the "ubiquitious Steppe_EMBA" comes from:
Image
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

Rudradev wrote: Is it surprising that their model got match hits with even 140 out of 246 Indian population samples?

Not at all. See, their model was Steppe_MLBA + Iran_N + AASI.

Steppe MLBA= 74% Steppe_EMBA and other things.

Steppe_EMBA = Iranian Chalcolithic and European Hunter-Gatherer.

Iranian Chalcolithic presumably descended at least in part from Iranian Neolithic (i.e. Iran_N, or Iranian agriculturists).

Iranian Agriculturists per the SAME paper share ancestry with ALL Indians, ASI, ANI, and the contrived "Indus Periphery".
Staring at the Laziridis charts for a bit.

Did Steppe_EMBA or Steppe_MLBA originate anything (e.g.,R1a)
a. that is not there in the (Iranian Neolithic/Chalcolithic/Agriculturalist + AASI + European Hunter-Gatherer + Europe_MNChl)
and
b. that is found "ubiquitously" in Indians?

If yes, then Steppe_EMBA or Steppe_MLBA as ancestors of Indians makes sense.
If no, then the whole Steppe thing is BS.
Rudradev
BRF Oldie
Posts: 4226
Joined: 06 Apr 2003 12:31

Re: Out-of-India - From Theory to Truth: Part 2

Post by Rudradev »

A_Gupta wrote:
Rudradev wrote: Is it surprising that their model got match hits with even 140 out of 246 Indian population samples?

Not at all. See, their model was Steppe_MLBA + Iran_N + AASI.

Steppe MLBA= 74% Steppe_EMBA and other things.

Steppe_EMBA = Iranian Chalcolithic and European Hunter-Gatherer.

Iranian Chalcolithic presumably descended at least in part from Iranian Neolithic (i.e. Iran_N, or Iranian agriculturists).

Iranian Agriculturists per the SAME paper share ancestry with ALL Indians, ASI, ANI, and the contrived "Indus Periphery".
Staring at the Laziridis charts for a bit.

Did Steppe_EMBA or Steppe_MLBA originate anything (e.g.,R1a)
a. that is not there in the (Iranian Neolithic/Chalcolithic/Agriculturalist + AASI + European Hunter-Gatherer + Europe_MNChl)
and
b. that is found "ubiquitously" in Indians?

If yes, then Steppe_EMBA or Steppe_MLBA as ancestors of Indians makes sense.
If no, then the whole Steppe thing is BS.
Arun, except in some rare cases it's virtually impossible to determine where a specific marker originated.

Time to Most Recent Common Ancestor for R1a, e.g. is ~24000 years. Way before any Bronze Age.

To really answer your question we would have to drill down into the Vagheesh et al data and look at their tables SNP by SNP. Only then will we know if there are markers absolutely unique to Steppe_MLBA, not found anywhere else, that show up in Indian populations to a significant degree.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

From the supplemental material in Lazarides 2016
https://www.biorxiv.org/content/biorxiv ... 9311-1.pdf
We next tried all possible Choice(20, 2)= 190 pairs of the 20 West Eurasian populations of Table S9.1, thus Left=(South Asian, West Eurasian1, West Eurasian2, Onge, Han). South Asian populations could not be modeled as mixtures of most of the 190 quadruples (pairs of West Eurasian populations plus Onge and Han). In Table S9.2 we list the 49 pairs which resulted in successful modeling of at least 4 South Asian populations (p-value for rank=3 greater or equal to 0.05 and estimated mixture proportions in [0, 1] interval). A clear pattern emerges with the most successful pairs involving a population from Iran/the Near East and one from the Steppe/Eastern Europe. The fact that the Eastern European population is either EHG, Steppe_Eneolithic, or Steppe_EMBA is not surprising, as Steppe_Eneolithic and Steppe_EMBA groups are themselves mixtures of EHG and Iran-related populations as we have seen in Supplementary Information, section 7.
In Table S9.3 we see that only 12 pairs of West Eurasian populations can be used to model at least 4 South Asian populations. All these pairs involve a steppe population and a population from Iran or CHG. Only steppe/Iran combinations can model all or nearly all South Asian populations
successfully. In Table S9.4 we list the inferred mixture proportions for the top 3 pairs that can be used to model nearly all South Asian populations in our dataset. Ancestry from both Iran and Steppe related sources is pervasive across South Asia. Estimated proportions of EHG ancestry are lower than Steppe_Eneolithic ancestry which are lower than Steppe_EMBA ancestry, consistent with the dilution of EHG ancestry from a southern source5 related to populations of Iran (Supplementary Information, section 7), thus more Steppe ancestry is required when the Steppe source has less EHG ancestry.
West Eurasian-related ANI ancestry in South Asia may pre-date, coincide with, or postdate Indo-European dispersals, although a partial link between the two is suggested by the evidence for Bronze Age admixture in India1 that contributed a large portion of ancestry especially in Indo-European speakers whose magnitude would be compatible with major linguistic change. However, ANI ancestry related to both ancient
Iran and the steppe is found across South Asia (Table S9.4) making it difficult to associate it strongly with any particular language family (Indo-European or otherwise).
{Shiv's point above} Nonetheless, the fact that we can reject West Eurasian population sources from Anatolia, mainland Europe, and the Levant diminishes the likelihood that these areas were sources of Indo-European (or other) languages in South Asia. While the Early/Middle Bronze Age ‘Yamnaya’-related group (Steppe_EMBA) is a good genetic match (together with Neolithic Iran) for ANI, the later Middle/Late Bronze Age steppe population (Steppe_MLBA) is not. {But the 2018 pre-print overthrows this}. Steppe_MLBA includes Sintashta and Andronovo populations who have been proposed as identical to or related to ancestral Indo-Iranians, as well as the Srubnaya from eastern Europe which are related to South Asians by their possession of Y-chromosome haplogroup R1a1a1b2-Z93
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

^^^
Is there any reason to prefer the left or right mixtures over the middle mixture? I suppose the smaller p-value is less likely?

Image
Rudradev
BRF Oldie
Posts: 4226
Joined: 06 Apr 2003 12:31

Re: Out-of-India - From Theory to Truth: Part 2

Post by Rudradev »

No, smaller p value indicates less chance of it being a random observation and more likelihood of your hypothesis being correct. Hence the right column is the best "fit".
Nilesh Oak
BRFite
Posts: 1670
Joined: 11 Aug 2016 06:14

Re: Out-of-India - From Theory to Truth: Part 2

Post by Nilesh Oak »

A simple formula to remember - p values...

If p< 0.05, Whatever it is is....(at 95% statistical confidence)....is significant...or is significantly different.

p <0.10 (90% statistical confidence) and so on...
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

^^^From the recent preprint, e.g, caption of table S4.1 "We highlight in green passing p-values (using a threshold of p>0.01)."
:)

Anyway, it is a key table in what Rudradev told us.

Image
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

Skanda wrote:Shiv,

Did you read this: https://twitter.com/anupampom/status/982284507293331456
No. Can't understand it..

In any case he fails on the same count. How do you connect genetics with language? Some one ask him and gradually the "evidence" starts falling apart..

Have you seen this?
https://twitter.com/bennedose/status/983514474442244096
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

Genetics does not code for langauge.

So how do genetics researchers assert that they have found the origin of a language?

They have not. They depend on linguists who made up non existent PIE and placed that non existent PIE in Steppe where no known language was spoken in that era. Unprovable geography+non existent language

Has anyone seen this: To be diligent one need to look at the original authors and references used to claim that a particular language was spoken in steppe region. In fact there is no evidence. No pottery, no tablet. No poem. No narrative. Nothing.Into this emptiness linguists have pasted "PIE"

Quoting myself:
They are saying that horse and wheel came with language to India. The proof "Steppe culture is same as Rig veda" who said this: David Anthony (widely quoted) claims that Kurgan graves in Steppe are mentioned in Rig Veda
This is how he claims that Steppe culture == Rig Veda.
https://pbs.twimg.com/media/DaRBweOVwAAknPD.jpg
Image

Finally may I point out:
Are Vedas lies? Are they fairy tales? If yes - then we must dismiss everything as rubbish - including mention of horses and chariots. If they are not lies then we must consider fast flowing Saraswati river to sea and archaeoastronomy references as true. Can't reject one and take the other.

You can't claim horse and language came in 1500 BC while Pre 2000 BC river and astronomy are fake.

but then again if you can swallow that you could be a linguist

Problem is we don't check the original references. We should.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

If languages do correlate with genes, and we believe in the antiquity of Sanskrit, then the solution has to be that it was the language of peoples on the AASI-Iran Chalcolithic cline, and entered the Steppe by the Chalcolithic Iranian ancestry of Steppe_EMBA.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

^^^ Also, AFAIK, Rg Veda 10:18:8 enters into discussions of Sati.

8. Rise, come unto the world of life, O woman: come, he is lifeless by whose side thou liest.
Wifehood with this thy husband was thy portion, who took thy hand and wooed thee as a lover.

e.g.,
http://koenraadelst.blogspot.com/2013/0 ... -sati.html

It is funny how the same sukta supposedly refers both to the funeral pyre and to the burial chamber :) :) :)
gandharva
BRF Oldie
Posts: 2304
Joined: 30 Jan 2008 23:22

Re: Out-of-India - From Theory to Truth: Part 2

Post by gandharva »

Prem Kumar wrote:Arun: that's what I am thinking too. But just because a particular mutation was found in the aDNA sample from the Steppes, doesn't mean its "Steppe DNA". This mutation could have very well evolved elsewhere and traveled to the Steppes.

Who's to say that the ancestor the Steppe DNA will not be found in aDNA from IVC?

The logic used by geneticists is:

1) Steppe aDNA has some mutation
2) This mutation is found in Indians too
3) So, Steppe people must have invaded India

The fact that there is no aDNA from India is conveniently ignored. If Indian aDNA is found to be more ancient & also have Steppe DNA, the same points (1) and (2) can be used to argue for OIT!
I asked this question to Vagheesh. Here is his reply. He believes that if aDNA is found, it's not likely to have Steppe DNA.

"This is not going to change. All the evidence is pointing clearly to the arrival of steppe ancestry in India to between 1500-2000BCE. Even if we do find aDNA in the Indus Valley, or indeed in the Deep South, as we state in the paper it is either Iran_N + AASI , or AASI"

https://twitter.com/vagheesh/status/982095419667697665
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

A_Gupta wrote:If languages do correlate with genes, and we believe in the antiquity of Sanskrit, then the solution has to be that it was the language of peoples on the AASI-Iran Chalcolithic cline, and entered the Steppe by the Chalcolithic Iranian ancestry of Steppe_EMBA.
Vagheesh asked me before everything descended into war mode "So what do I think needs to be done?'. My reply was that we have to look at migrations and movements that occurred far faar earlier. This is my pet peeve. Linguists and Archaeologists have geneticists by the balls because they have already ascribed fake dates to everything and geneticists are always looking for some "useful and interesting" correlation that fires people up. Finding obscure genetic diseases is no good. So they pick up texts with fake stories of migration, match them to some findings, fit the data if need be & and simply quote the linguistic sources without looking deeper - making the latter "true" by association with gene papers.

After ehated discussions with Vagheesh he said "OK we will correlate language with places that have textual proof first, archaeology next and finally linguistic evidence - and claimed that all 3 were true for stepp migration to Europe. I was out of the discussion by then and did not feel like jumping back in to point out that the fake PIE in steppe is hardly "proof" of source of language - so all his work may end up GIGO anyway.

I just jumped in today about a point made by Rudradev who said that exclusion of certain Indian populations is clearly cherry picking something to fit teh data. Vagheesh replied that the "excluded" people had Tibeto-Burman ancestry like Bengali speakers and Austro Asiatic speakers. But duuuh! tell me this - if Steppe genes brought IE language why the hell are you excluding Bengali speakers - that language is chock full of of Sanskrit. Coincidentally racist Brits (I have the cite) decided to exclude Indians from their "Aryans" because they did not want to be associated with the dark skinned Bengalee.

Hey maybe I will take a screenshot and post on Twitter and muddy the discussion...
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

OK I gone and dun it
https://twitter.com/bennedose/status/983712522787143680
Sorry to say this - maybe this was unintended, but excluding Bengali speakers sounds eerily like what a Briton called L. Huxley said about "Bengalees" to exclude them from the "pure Aryan language speakers". Your selection may have inadvertently done that:
Image
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

gandharva wrote:"This is not going to change. All the evidence is pointing clearly to the arrival of steppe ancestry in India to between 1500-2000BCE. Even if we do find aDNA in the Indus Valley, or indeed in the Deep South, as we state in the paper it is either Iran_N + AASI , or AASI"
I'm dubious that Rakhigarhi aDNA will ever be published - one would imagine if it existed, there would be great urgency to publish it, but the above means this pre-print will be falsified if Rakhigarhi aDNA has anything more than Iran_N + AASI, it doesn't matter what that extra is.

My sort-of-unjustified-hope is that the folks that supposedly have the Rakhigarhi aDNA wanted the Steppe-invasionists' folks to put a definite stake in the ground, so that they cannoteasily weasel out of their position; and that is why they've held back on publication. But I "know" that is a false hope.
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

Q: since the "Indus_Periphery" have no Steppe ancestry, but are made of Iran_N + AASI; but Steppe_EMBA itself is Iran_Chacolithic + European_Hunter_Gatherer (EHG), is the basic statement that "Indus_Periphery" has no EHG component?)

Also notice some counter-intuitive things - Indians were modeled based on two hypothetical ancestral populations that were distinguishable - ANI and ASI. Now we are told that both ANI and ASI are descended from various mixtures of (Indus_Periphery + Steppe + AASI).

2009:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2842210/
One, the “Ancestral North Indians” (ANI), is genetically close to Middle Easterners, Central Asians, and Europeans, while the other, the “Ancestral South Indians” (ASI), is as distinct from ANI and East Asians as they are from each other.
Remember that ANI, ASI are theoretical constructs. In 2009, East Asians, ANI, ASI were all equally distinct. Now ANI, ASI are the common descendants of (Indus_Periphery + Steppe + AASI). Surely the definitions of ANI, ASI are as they were in 2009, as theoretical constructs, not backed by any explicit aDNA? My "math spider sense" is tingling that something is wrong in the geometry of a very-high dimensional space. I think I will have to invest months if not years in a series of genomics classes to get myself straight, until I understand the algorithms used and know what they're really talking about.

That is, if the current preprint is correct, it should have been possible to take the 2009 data, and postulate a ghost population, so that ANI, ASI lie on a cline of this ghost population and AASI, in the same way Steppe_EMBA lies between Iran_Chcl and EHG. Cline is a simple one-dimensional construct; I really mean the two-dimensional construct. The genomics algorithms don't seem to reveal structure, rather only confirm/deny structure that you put in as your hypothesis.
Rudradev
BRF Oldie
Posts: 4226
Joined: 06 Apr 2003 12:31

Re: Out-of-India - From Theory to Truth: Part 2

Post by Rudradev »

Arun,

As introductory material I found both these youtube video courses quite engaging and well-made.

This is a series of classroom lectures about introductory bioinformatics and the analytical techniques commonly used. It seems geared more to people with a comp-sci or math background than to biologists. From UC Davis.

https://www.youtube.com/watch?v=liNblw4 ... VKCGCDmmoL

This one is about molecular evolution, from Coursera. It gets to a lot of the things we're talking about from a fairly simple starting point in biology.

https://www.youtube.com/watch?v=okjVaLA ... GGckG0sG5y
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

I had done the first course of this 8-course Coursera series over an year ago, but then decided not to continue, largely because I was short of time.
https://www.coursera.org/specialization ... ta-science
Wondering if I should pick up this series or some other?
Prem Kumar
BRF Oldie
Posts: 4218
Joined: 31 Mar 2009 00:10

Re: Out-of-India - From Theory to Truth: Part 2

Post by Prem Kumar »

Rudradev: I looked at the Lazaridis paper of 2016 that you pointed out. A couple of questions arose:

1) If they actually have Steppe_EMBA aDNA, why do they need to "model it"? Is it so that they can then take that model and apply it elsewhere?

2) Aren't the clusters used to "model" Steppe_EMBA arbitrary & based on whatever aDNA is available? Example: why not model Steppe_EMBA on a cline between Iran_ChL and Indus_Valley_Cluster?

3) Does the Indus_Valley_Cluster not make an appearance here because we have no aDNA from there? If so, modeling Steppe_EMBA based on only "available aDNA" (i.e. CHG, EHG, Iran etc) is like a drunk searching for his lost keys under the streetlamp because "that's where the light is"!

4) How do they know the "directionality"? Example: in Extended Data Figure 4, Mala is claimed to have Steppe_EMBA, Iran & Onge genes. Couldn't it just as likely be the opposite direction: i.e. Mala's ancestors spread out and gave rise to Iran and Steppe_EMBA genes?
Rudradev
BRF Oldie
Posts: 4226
Joined: 06 Apr 2003 12:31

Re: Out-of-India - From Theory to Truth: Part 2

Post by Rudradev »

1. To estimate what the ancestry of Steppe_EMBA is. You use clusters determined from presumably older sample populations (Chalolithic, Hunter-Gatherer) to see if the resultant admixture can account for ancestry of a later population (Early-Middle Bronze Age).

2. Sure, somebody has to do it and publish the results of statistically testing such a model, that's all! It was not in the interest of Vagheesh et al to test that model for reasons you can imagine (the fraud David Anthony is one of the co-authors, so he probably steered them away from testing anything that could cast doubt on his AIT theorizing).

Another issue in what you propose is the supposed antiquity of the Indus_Periphery cluster. If it is bronze age, then it is possibly not ancestral to Steppe_EMBA ("Early-Middle Bronze Age"). We need earlier-than-bronze-age samples of Indian aDNA to do a test such as you propose.

3. The so-called Indus_Periphery cluster... if indeed those three individuals ARE from Indus Valley (which is just a postulate), was sampled only after the Lazaridis paper came out. The aDNA from those three individuals was not available to Lazaridis et al. Why Vagheesh et al didn't test it is anybody's guess.

4. Mala is a modern Indian population while Steppe_EMBA and Iranian_Agriculturists are aDNA populations. Many assumptions are made (e.g. that Onge, also a modern population, is a nearly pure form of what the authors call "AASI"... again, a completely theoretical construct, as described in earlier posts).

The way directionality of a cline is traditionally established is to consider the frequency of markers that occur at each end of the cline. So for example, if ASI markers are more common in Mala (modern southern Indian) and less common in Kashmiris (modern northern Indian), the assumption is made that ASI markers originated in the south and spread from south to north. This for example is what the Moorjani et al paper says. Then that hypothesis is tested against what aDNA samples have to reveal, which is what Vagheesh et al purport to be doing... although they go about their testing in exquisitely selective ways.

Ultimately without aDNA from India it's very hard to make the argument that this happened. We have a "blank" in terms of Indian-origin markers across the board. The likes of Vagheesh feel emboldened to fill those blanks in for us.
RoyG
BRF Oldie
Posts: 5620
Joined: 10 Aug 2009 05:10

Re: Out-of-India - From Theory to Truth: Part 2

Post by RoyG »

A_Gupta wrote:
gandharva wrote:"This is not going to change. All the evidence is pointing clearly to the arrival of steppe ancestry in India to between 1500-2000BCE. Even if we do find aDNA in the Indus Valley, or indeed in the Deep South, as we state in the paper it is either Iran_N + AASI , or AASI"
I'm dubious that Rakhigarhi aDNA will ever be published - one would imagine if it existed, there would be great urgency to publish it, but the above means this pre-print will be falsified if Rakhigarhi aDNA has anything more than Iran_N + AASI, it doesn't matter what that extra is.

My sort-of-unjustified-hope is that the folks that supposedly have the Rakhigarhi aDNA wanted the Steppe-invasionists' folks to put a definite stake in the ground, so that they cannoteasily weasel out of their position; and that is why they've held back on publication. But I "know" that is a false hope.
Are they hiding something?
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

The traditionally priestly castes per Narasimhan et. al.
Pandit
Brahmin_Haryana
Brahmin_Nepal
Bhumihar_Bihar
Brahmin_Tiwari
Brahmin_UP
Bhumihar_UP
Havik
Brahmin_Karnataka
Brahmin_Bhatt
Brahmin_Vaidik
Brahmin_Catholic_Goa
Brahmin_Catholic_Mangalore
Brahmin_Catholic
Brahmin_Catholic_Kumta
Gaud_Karnataka
Vishwabrahmin
Brahmin_Uttrakhand
A_Gupta
BRF Oldie
Posts: 12065
Joined: 23 Oct 2001 11:31
Contact:

Re: Out-of-India - From Theory to Truth: Part 2

Post by A_Gupta »

What to make of this comment on biorxiv? The idea is that the Steppe component of Indian samples can be attributed to the Iranian Neolithic.
http://disq.us/p/1rj4yuo
Maju • 5 days ago
I'm sorry but, while the amount of data revealed here is very impressive, the conclusions re. steppe influence do not seem to match with the data in any obvious way.

First figure 2 (because it's easier): All steppe populations (except Dali and Okunevo) have sizable EHG component (light blue) but not a single one of the Iran-Turan South Asia have it at all. This basically precludes any significant steppe-originated admixture south of the steppe, unless you'd claim Dali as the origin, which I do not see stated anywhere (Okunevo would not work either because of their East Asian component).

Second fig. 1D: the main steppe component is here colored orange and it does show some presence in Iran-Turan since the Chalcolithic, long before any sort of steppe admixture would be expected, and you seem to agree that the area is only peripheraly influenced by this inflow of steppe ancestry and otherwise remains the same. However this also implies that the presence of this orange component in the South Asia samples from Northern Pakistan can be fully or almost fully attributed to Neolithic influence from Iran.

I'm not saying that your model is not ultimately correct, I'm probably missing something, but it is not in any way obvious, striking, clear and hard to refute. And also seems to be something much more subtle than the phrasing used in the conclusions seems to suggest. So I would suggest that you'd work in the direction of making your conclusions more clear, because they seem to rely ONLY on qpAdm (fig. 3) results which are not only low-confidence (only one statistical analysis algorithm seems to say so, while the rest do not in any obvious way) but also not easy to understand.

Disclaimer: I am strongly favorable to the Kurgan model of Indoeuropean expansion, although I would allow for negligible genetic admixture in the Metal Ages associated to cultural-linguistic expansion via elite domination (not only in India but also in Western Europe maybe, just as we see no or very minor genetic influence in the processes of acculturation in Turkey and Hungary, or in the Slavic Balcans). In this sense at least I don't think I'm biased against your model, and I'd be happy to find at least some minor genetic signal from the steppe influencing Southern Asia. Just that I do not see it or at most I see a very marginal and not in any way obvious fit. Maybe you should tone down your conclusions? The data is still massive and massively interesting.
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

RoyG wrote:
A_Gupta wrote:
I'm dubious that Rakhigarhi aDNA will ever be published - one would imagine if it existed, there would be great urgency to publish it, but the above means this pre-print will be falsified if Rakhigarhi aDNA has anything more than Iran_N + AASI, it doesn't matter what that extra is.

My sort-of-unjustified-hope is that the folks that supposedly have the Rakhigarhi aDNA wanted the Steppe-invasionists' folks to put a definite stake in the ground, so that they cannoteasily weasel out of their position; and that is why they've held back on publication. But I "know" that is a false hope.
Are they hiding something?
Unlikely. Here is what I think

1. Very little or almost no DNA retrieved - I am not sure of the numerical value but I think genetics researchers look for sequences of DNA that have at least 200,000 "SNP"s (out of a total of many million that can be there in a complete human DNA set) That number is assumed to be large enough to do a comparison with other humans. That is the number chosen by Vagheesh and co I think.
2. DNA is contaminated - with goat/boar/bacterial/human DNA

Rakhigarhi DNA would be huge - they would not hide it because it will place the researchers on a high genetic pedestal and give them free trips to conferences to present their findings
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

Let me ask all you wise people here a simple question. I just wonder why so few are asking this question.

Let us say the "Kurgan" model of language spread is right. A Kurgan is a type of grave. What is the connection between Kurgan and India? There are no Kurgans in India. (*- see footnote). Why does no Indian ask how Russian graves have been connected with India? I have frequently suggested the answer - but still people keep talking about a Kurgan connection with steppe as if it is within the realms of possibility. What is the connection between Kurgan and India. Why do we even consider it as possible?

Footnote: The Vagheesh Narasimhan paper has a detailed account of the structure of steppe graves and Gandharva of BRF earlier posted a paper about the structure of IVC graves. I have copypasted that stuff on to a document that I will post/publish in due course.
Anshuman.Kumar
BRFite -Trainee
Posts: 58
Joined: 08 Sep 2016 20:16

Re: Out-of-India - From Theory to Truth: Part 2

Post by Anshuman.Kumar »

Prem Kumar wrote:Had an exchange with Dr. Kumaraswamy Thangaraj. He said Rakhigarhi DNA publication would take several months. So, those holding their breath, please don't. The article in Chindu by Tony Joseph earlier this year (that a paper was due soon), was a lie - just like other lies by him and the Chindu.

This is what guess we must keep in mind when talking about rakhi garhi DNA.

And it looks likely there was a sabotage attempt in foreign labs..So no wonder birbal sahni institute was pressed into service
Prem Kumar
BRF Oldie
Posts: 4218
Joined: 31 Mar 2009 00:10

Re: Out-of-India - From Theory to Truth: Part 2

Post by Prem Kumar »

Looks like they don't have good data on Rakhigarhi samples. I don't know if this means that they are trying (or) they have tried and given up.

This is very unfortunate. The earlier set of Rakhigarhi DNA could not be analyzed due to contamination. There was hope by the researchers that the new set would yield results (because they had better protocols to prevent contamination and better labs to analyze the results). But things don't seem to have improved much

Rudradev: I understand that Mala is an existing Indian group and is not "ancestral'. But wouldn't DNA analysis help determine if there are present day Mala who carry ancient genes which could be predecessors to the Steppe/Iran markers that they also carry? IOW, is aDNA the only way?
Prem Kumar
BRF Oldie
Posts: 4218
Joined: 31 Mar 2009 00:10

Re: Out-of-India - From Theory to Truth: Part 2

Post by Prem Kumar »

A_Gupta wrote:What to make of this comment on biorxiv? The idea is that the Steppe component of Indian samples can be attributed to the Iranian Neolithic.
http://disq.us/p/1rj4yuo
I had posted the same comment on this forum as well. I found it very incisive. It looks like the data doesn't support the paper's conclusions.

Perhaps someone like Rudradev can take a stab at it!
shiv
BRF Oldie
Posts: 34982
Joined: 01 Jan 1970 05:30
Location: Pindliyon ka Gooda

Re: Out-of-India - From Theory to Truth: Part 2

Post by shiv »

https://indo-european.eu/2017/09/marija ... -cultures/
According to Gimbutas, the “Kurgan people” are evidenced by single graves in deep shafts, often in wooden chests (coffins) or stone cists marked by low earth or stone barrows; the dead lay on their backs with legs contracted; they were buried with flint points or arrowheads, figurines depicting horses’ heads, boars tusk ornaments and animal tooth pendants. Human sacrifice was allegedly performed during the funeral ceremonies,and sometimes ritual graves of cattle and other animals were added. This is said to contrast with what Gimbutas called the culture of Old Europe (i.e. the earlier Neolithic of the Balkans), who “betray a concern for the deification of the dead and the construction of monumental works of architecture visible in mortuary houses,grave markings, tumuli, stone rings or stone stelae, and in the large quantity of weapons found in the graves”.
Anshuman.Kumar
BRFite -Trainee
Posts: 58
Joined: 08 Sep 2016 20:16

Re: Out-of-India - From Theory to Truth: Part 2

Post by Anshuman.Kumar »

Prem Kumar wrote:Looks like they don't have good data on Rakhigarhi samples. I don't know if this means that they are trying (or) they have tried and given up.

This is very unfortunate. The earlier set of Rakhigarhi DNA could not be analyzed due to contamination. There was hope by the researchers that the new set would yield results (because they had better protocols to prevent contamination and better labs to analyze the results). But things don't seem to have improved much

Rudradev: I understand that Mala is an existing Indian group and is not "ancestral'. But wouldn't DNA analysis help determine if there are present day Mala who carry ancient genes which could be predecessors to the Steppe/Iran markers that they also carry? IOW, is aDNA the only way?

The Dainik jagran article and earlier article by Dubey/Thangraj in Hindu as response to Joseph clearly said that they(Rai etc) have data,
The book by Reich,this paper with a specifically chosen person as lead author with Force fit data to suit specifically chosen conclusions gives a clear indication that they have tried their best to sabotage/Discredit Rakhi garhi results..
If you can get dNA from.swat,Iran you can get from Rakhi garhi too..
Anshuman.Kumar
BRFite -Trainee
Posts: 58
Joined: 08 Sep 2016 20:16

Re: Out-of-India - From Theory to Truth: Part 2

Post by Anshuman.Kumar »

"Moreover, there is evidence which is consistent with the early presence of several R1a branches in India (our unpublished data)."

Etc etc

https://www.google.co.in/amp/www.thehin ... 7.ece/amp/

Just because we have a pre print from a Harvard Person does not mean everything is settled..

The surfeit of articles etc was a clear indication that the Western Scholars were ready with a contrived paper and in a good position to muddy waters for Rakhi garhi results.
Prem Kumar
BRF Oldie
Posts: 4218
Joined: 31 Mar 2009 00:10

Re: Out-of-India - From Theory to Truth: Part 2

Post by Prem Kumar »

Good points, Anshuman. I'd forgotten that they'd talked about unpublished data. But I heard that Rakhigarhi paper was several months away, at least.

Its interesting to note that K. Thangaraj (who'd be involved in any Rakhigarhi paper) is also a co-author of the Vagheesh paper, though he is not a lead author. He said they didn't include Rakhigarhi data in the Vagheesh paper because good data wasn't available. Don't know what that means......
Post Reply