Population genomics of Bronze Age Eurasia (Allentoft et al. 2015)

[Humanist]

The poster Wojewoda mentioned this on another forum. He compared the K9 values for the RISE samples from Remedello and modern Italians. The Remedello Italians are 100% "MN_European."

SE_Asian Siberian South_Central_Asian East_African San_Bushman Sub-Saharan Steppe Middle_Eastern MN_European

0%	0%	0%	0%	0%	0%	30%	23%	46%	Italian_Bergamo
0%	0%	4%	2%	0%	0%	21%	39%	34%	Italian_EastSicilian
0%	0%	1%	1%	0%	0%	22%	42%	33%	Italian_South
0%	0%	1%	0%	0%	0%	28%	30%	40%	Italian_Tuscan
0%	0%	2%	2%	0%	1%	23%	39%	35%	Italian_WestSicilian
0%	0%	0%	0%	0%	0%	0%	0%	100%	RISE487
0%	0%	0%	0%	0%	0%	0%	0%	100%	RISE489

[Moderator]

[MOD] Off-topic regarding Scythians and Saka split to here.

[paulgill]

Actually what Krefter is saying makes perfect sense. Androvono/ Sintashta people are hardly invaders more like wandering pastoralists.

If that is right then we need to go no further for ancient DNA evidence than the known ancient sites of killing fields, created by Genghis Khan, Atilla the Hun, etc. . There should be tons of it available where virtually everyone in some cities was killed by these Devil Incarnates, NO?

[Hando]

I've been gone all day, and a lot seems to have happened in terms of discussion.

To address a few points:

4) WHG as understood in Lazaridis et al. has a very different meaning now, post-Haak et al. We now know that WHGs were probably admixed between North Eurasian people and some sort of populational group that was basal to North Eurasian people (but they were certainly a stabilized mixture). EHG and MA1 are different spatio-temporal nodes on the "North Eurasian" populational network. The befuddlement a lot of people here are feeling is mainly because they still have the "old model" in mind. The WHG-ANE-ENF model wasn't final.

Are you saying that WHG is not simply descended from Gravettians, Magdalenians, but are mixed with another population? And what exactly is "basal to North Eurasian? Is that ancestral to ANE, WHG and EHG?

[vettor]

Thank you. Well, that is certainly interesting, given that, as mentioned previously, the skeleton dates to the same period as Mitanni. Not that they are necessarily connected, but it is interesting nonetheless.

more on swastika ............part of the vinca script ancient alphabet

http://www.omniglot.com/writing/vinca.htm

[Kale]

I see 11/24 runes, and 16/26 capital English letters in that script. English letters are a lot simpler though, C, V, W, M, X, etc. Whereas that Vinca script has the oddballs of Pertra, numerous that are close to Jara, as well as multiple versions of Odal.

[Michał]

RISE392 is 7.85% older than indicated by the YFull’s estimate for Y877.
RISE386 is 11.9% older than indicated by the YFull’s estimate for YP1460.

And here are two more examples for R1a:

RISE446 is 17.3% older than indicated by the YFull’s estimate for S3479.
RISE61 is 21.0% older than indicated by the YFull’s estimate for Z281.

One of our forum members has made me aware of a small error in my above calculations, so here are the corrected percentages:

RISE392 is 8.52% older than indicated by the YFull’s estimate for Y877.
RISE386 is 13.5% older than indicated by the YFull’s estimate for YP1460.
RISE446 is 20.3% older than indicated by the YFull’s estimate for S3479.
RISE61 is 27.2% older than indicated by the YFull’s estimate for Z281.

[TigerMW]

One of our forum members has made me aware of a small error in my above calculations, so here are the corrected percentages:

RISE392 is 8.52% older than indicated by the YFull’s estimate for Y877.
RISE386 is 13.5% older than indicated by the YFull’s estimate for YP1460.
RISE446 is 20.3% older than indicated by the YFull’s estimate for S3479.
RISE61 is 27.2% older than indicated by the YFull’s estimate for Z281.

Thank you. I assume you are talking about the estimated "formed" date, right?
Have you looked at completely different ancient haplogroups as well? Are YFull's formed estimates consistently low or is this sporadic by haplogroup?

[Michał]

Thank you. I assume you are talking about the estimated "formed" date, right?

This is more or less correct. More precisely, this is a "formation" age for a given clade (as estimated by YFull) minus a relatively small number of years that account for at least one positive SNP in a downstream block of SNPs. So if the downstream block consists of let's say 10 mutations, while the difference between the formation age and the expansion age is 500 years, I am deducting 50 years from the formation age. Using this method, I slightly overestimate the YFull age for a given SNP, because we don't know the exact position of a given SNP in a sequence of sub-sequentially arising mutations, but this is certainly a better option than using the formation age only.

Have you looked at completely different ancient haplogroups as well? Are YFull's formed estimates consistently low or is this sporadic by haplogroup?

I have just posted my calculations for sample RISE98 from clade R1b-U106, and the results seem to confirm my initial conclusion based on the R1a samples. BTW, I see no formal reasons why the SNP rate in members of a given haplogroup/clade should be significantly different from that in other haplogroups/clades. We can expect to see such a difference when comparing all SNPs downstream of M173 in R1a vs R1b members but this is because both these haplogroups expanded relatively recently, so most SNPs downstream of M173 are those that arose in a single lineage (for example in a lineage common to all M417 members or in a lineage common to all L23 members), which makes such numbers prone to deviate from the average. However, there is no data suggesting (nor any serious reason to suspect) that any number of independent R1a lineages will show a statistically different mutation rate than a similar group of independent R1b lineages.

[TigerMW]

... BTW, I see no formal reasons why the SNP rate in members of a given haplogroup/clade should be significantly different from that in other haplogroups/clades. We can expect to see such a difference when comparing all SNPs downstream of M173 in R1a vs R1b members but this is because both these haplogroups expanded relatively recently, so most SNPs downstream of M173 are those that arose in a single lineage (for example in a lineage common to all M417 members or in a lineage common to all L23 members), which makes such numbers prone to deviate from the average. However, there is no data suggesting (nor any serious reason to suspect) that any number of independent R1a lineages will show a statistically different mutation rate than a similar group of independent R1b lineages.

I certainly don't pretend to understand the workings of the Y chromosome, but I have observed (anecdotally) that that there are individuals within 2500 year old (or so we think) haplogroups that have significantly different (up to 30%) difference in SNP counts, even after weeding to get down to "safe" SNPs.

Since I have seen those occurrences, I ask from the opposite perspective. I think it is incumbent on those using SNP counting to estimate ages across haplogroups to show the evidence that SNP branch lengths have a precise linear relationship with time. A 30% difference over 3000 years could be cause for alarm of wide variability. There is a big difference between honing in on a 200-300 year period 4000 years ago versus honing versus honing in on the correct millennium.

Again, I still think SNP age estimation methods are very valuable, but the precision is a question mark. I haven't seen the scientific studies and rebuttals like I have with Y STR methods yet. I would consider that we have initial studies but not rebuttals and/or broad agreement.