Medieval social landscape through the genetic history of Cambridgeshire before and af

[Michalis Moriakos]

Hello, if you were going to ask if I'd found out about the unattributed samples, then the answer is obviously no, so I'm going to contact the authors and if I get a reply I'll share the details here.

Hey, I figured out what those samples are pretty quickly now that the official paper has been released.

[Capitalis]

Hello, if you were going to ask if I'd found out about the unattributed samples, then the answer is obviously no, so I'm going to contact the authors and if I get a reply I'll share the details here.

Hey, I figured out what those samples are pretty quickly now that the official paper has been released.

Can you share?

Edit: No need. :-)

[Capitalis]

There are Norman-like samples from the pre-Norman period in the Gretzinger sample, with Y-DNA diversity.

I looked up the 'Norman-like samples' in Supplementary Table 2: Pairwise IBD sharing in Ringbauer et al.'s Accurate detection of identity-by-descent segments in human ancient DNA (2023). Only some of them were included for IBD analysis, and although the matches are interesting, such as sample I20644 IBD matching contemporaneous samples from Alt-Inden and Szólád, the inclusion of a clearly erroneous IBD match for VK150 with a Bronze Age Croatia sample that lived 2000 years prior to the Viking Age causes me to be doubtful about the IBD accuracy.

[Capitalis]

A very small number of Gretzinger et al.'s samples (e.g. I14538; 500-700 CE; Rookery Hill, Sussex) had a 'southeastern' profile, but not quite like the 'Austrasian' profile that I had been hoping to find (i.e. less southern, more eastern).

Genetic history of Cambridgeshire before and after the Black Death (2024) does have such samples, particularly 915 and 918; 940-1170 CE; Cherry Hinton, Cambridgeshire.

Thanks to guidance from one of the paper's authors on how to use a dropdown menu (!) at the ENA website - which to be fair to me, I'd never visited - I've located the sample IDs and updated the PCA plot with complete sample data.



I see ~900 years of genetic continuity in Cambridgeshire, from (potentially) the pre-Norman Age to the Victorian Age. :-)

[Capitalis]

This isn't a perfect overlay of this paper's UMAP plot of PoBI data with a PCA plot of G25 data, but it seems a reasonable effort, with caveats.



I began by overlaying PoBI Norfolk / Suffolk onto the East Anglian (pka Norfolk) family, and PoBI Northern Ireland onto Owen, then rotating a bit so PoBI Staffordshire was parallel with Molfish's mother.

As a result, we can draw a line between the West Country* (Devon [off the edge of the plot], Dorset and Somerset) and Southeast Wales (Gwent) and find JonikW, and we can draw a line between the West Country and what I'll call "core Ireland" (FionnSneachta, Jessie, Molfish's father) and find Capitalis.

Seabass' father plots halfway between his known ancestral groups (assuming far Northern England), and Molfish plots halfway between his parents in a little triangle, as do I with mine and Rodoorn (pka Finn) with his. This gives confidence that G25 is isolating specific ancestry.

So far, so good. Now for the bad:

* Lana plots as far more Irish than we would expect.
* Greg's position makes no sense relative to his ancestry.
* Scotland and Ireland aren't as stretched away from England on a G25 PCA plot as they are on this paper's UMAP plot. This isn't really "bad", per se.

Some unknowns:

* What do the "non-core" G25 Irish samples heading towards Wales represent? More Irish G25 coordinates with known ancestry would be useful; I used to have Garvan's but have lost them.
* Are some of the G25 Irish and Scottish samples crypto-English?
* Are the G25 Welsh samples "old" or "new" Welsh (I suspect the latter).

In reality, I think the UMAP plot is compressing the spread of some PoBI samples and exaggerating some others, so don't take my overlay too literally; e.g. the G25 Cornish samples plot around JonikW, rather than off to the left.

If you have G25 coordinates and know your Isles ancestry, I'll include you in any future plots if you private message them to me.

*I suspect the labelled Gloucestershire / Worcestershire clusters represent a subset of PoBI samples with a good amount of "old" East / Southeast Welsh ancestry, as there is a visible cline of Gloucestershire / Worcestershire samples running from Somerset to Gwent.

:-)

Edit: Forgot to add that PC's 1 and 2 contain ~71% of the variance, although I was interested to recently read Iosif Lazaridis quoting Nick Patterson's view that there is no need to label variance on PCA plots.