hringbauer.bsky.social
Population Geneticist | Ancient DNA
Research Group Leader at MPI-EVA Leipzig
www.hringbauer.com
53 posts
720 followers
494 following
Regular Contributor
Active Commenter
comment in response to
post
Yeah, the North America parallel came to our mind too! It's a great example of culture&language being not tied to one's genetic ancestry.
comment in response to
post
Thanks @aylwyn-scally.bsky.social and @pontus-skoglund.bsky.social 🙏
Initially, we had more Sicilian outliers genetically from the Eastern Mediterranean, but C14 dating always put them to Roman times (see PCA above - a interesting substory).
Seeing the Punic cline "emerge" was quite exciting!
comment in response to
post
Indeed, people of cental Mediterranean genetic origin adopted Levantine Phoenician culture & language. ✅️
As we lack aDNA from very early periods of Punic sites (when cremation dominated - which destroys DNA), we can only speculate on the exact process. My own bet is ongoing dynamic integration.
comment in response to
post
Marseilles was a Greek polis. 🏺
Our article focuses on aDNA from Punic sites in the Mediterranean (which, well, happen to have many people with Aegean-like ancestry too :)).
Studying aDNA from Greek colonies/polis would be important future aDNA work!
comment in response to
post
Thanks to a international interdisciplinary team behind this work! 🙏
A special shout-out to Ilan Gronau for his tireless work and the countless intriguing calls we had throughout.
And of course to my Postdoc supervisor, David Reich and his lab, who enabled this fascinating aDNA work! (4/4)
comment in response to
post
We find diverse ancestry in Punic sites 600-200 BCE - with a main component of Greek-Sicilian-like ancestry. Also North African ancestry was widespread across Punic sites on a narrow admixture cline - but remains a minority component (including in Carthage). (3/4)
comment in response to
post
We could generate genome-wide data of >100 human remains from prominent Punic necropolis in Iberia, Tunisia, Sicily, and the islands of Ibiza and Sardinia. That includes important urban centers, such as the former Phoenician colony Carthage itself. (2/4)
comment in response to
post
A fantastic international team is behind this work, including from @mpi-eva-leipzig.bsky.social.
Many thanks to everyone; in particular to the lead author @nadasalem.bsky.social who truly drove all this analysis and @arevsumer.bsky.social 's crucial contributions. 🙏
(7/7)
comment in response to
post
Finally, we detect only modest amounts of parental relatedness (measured by so-called ROH). We fit an effective size Ne~1,000. That is a considerable recent mating pool, not a crashing population!
Notably, the ROH also allowed us to quantify a lack of contaminating DNA by other humans. (6/7)
comment in response to
post
Previous ideas suggested that pastoralism spread into the region with the movement of people, such as from the Levant, NE Africa, or maybe even Europe.
Instead, our results now show that a largely local population adopted animal husbandry. 🐂📍 (5/7)
comment in response to
post
There are only a few Neandertal segments—around 10x less than out-of-Africa populations. This supports the idea that most ancestors of this lineage were not part of the out-of-Africa population that mixed with Neandertals, but rather stayed in Africa. (4/7) 🌍
comment in response to
post
Their ancestry resembles that of other ancient North Africans, including the 15ky-old Tafforalt group from Morocco. However, the Takakori individuals are much less admixed with Levantine-like ancestry.
This appears to be a deep lineage, previously undescribed in that form. (3/7)
comment in response to
post
Deep in the SW Libyan desert, the Takakori rock shelter preserved human remains and, despite the extreme climatic conditions, even small amounts of human DNA. Cutting-edge archaeogenetic methods let us squeeze out enough sequences from two individuals for various population genetic analyses. (2/7)
comment in response to
post
So I am afraid you are off by a factor ~3× too recent in the timing indeed 🙈
That also maps we'll to my experience in IBD segments sharing in aDNA - <10cm IBD segments mostly link samples >1000 years apart.
But it does not affect the really cool part of your amazing study - the sheer scale :)
comment in response to
post
I am with Shai here @aylwyn-scally.bsky.social
For older MRCAs more recombinations spots can start an IBD (increasing their total number). Your model does not account for this bias.
There is a history of analytical theory (perfectly fitting the sim.), e.g. SI here:
www.nature.com/articles/s41...
comment in response to
post
Yes you definitely get meaningful signal down to 2-4cM! ✅️ But is it still accurate IBD segments rather than some similarity measure? But that does matter mostly only for explicit modelling (e.g. timing).
Ralph&Coop did a careful copying in true IBD and checking their inference - a great plot:
comment in response to
post
Wait - detecting longer IBD segments should be much, much easier (higher power and lower FP) - modulo some gap merging, perhaps.
<6-8cm is where the trouble with lower power and higher error rates should start.
See e.g. the IBIS IBD caller paper
www.sciencedirect.com/science/arti...
comment in response to
post
Nice - Berlin is such a great place to stay and live!
I hope you can drop by to visit us at MPI-EVA in Leipzig - it is just 90min away.
comment in response to
post
ARG-inspired. 😀 Interesting that they claim full ARG inference is not too useful for compression on slightly noisy genomic data.
comment in response to
post
In other contexts, such as in the search for rare, deleterious variants, an emphasis on ancestry groups is unwarranted. In particular, while rare nucleotide variants are often described as “ancestry-specific”, they will in fact be found in large enough samples of any ancestry.
comment in response to
post
So what happened to ARG-based compression? 🤔I think that was even a main reason to develop scaleable ARG inference (tsinfer)