Updates to ASoSaH: new maps, updated PCA, and added newest research papers


The title says it all. I have used some free time to update the series A Song of Sheep and Horses:

I basically added information from the latest papers published, which (luckily enough for me) haven’t been too many, and I have added images to illustrate certain sections.

I have updated the PCAs by including North Caucasus samples from Wang et al. (2018), whose position I could only infer for older versions from previously published PCA graphs.

PCA of ancient and modern Eurasian samples. Early Eneolithic admixture events in the steppe drawn.

I have also added to the supplementary materials the “Tip of the Iceberg” R1b tree by Mike Walsh from the FTDNA R1b group, with permission, because some relevant genetic sections are centered on the evolution of R1b lineages, and the reader can get easily lost with so many subclades.

I have also updated maps, including some of the Y-DNA ones, and managed to finish two new maps I was working on, and I added them to the supplementary materials and to the menu above:

One on Yamna kurgans in Hungary, coupled with contemporaneous sites of Baden-Boleráz or Kostolac cultures:

Map of attested Yamnaya pit-grave burials in the Hungarian plains; superimposed in shades of blue are common areas covered by floods before the extensive controls imposed in the 19th century; in orange, cumulative thickness of sand, unfavourable loamy sand layer. Marked are settlements/findings of Boleráz (ca. 3500 BC on), Baden (until ca. 2800 BC), Kostolac (precise dates unknown), and Yamna kurgans (from ca. 3100/3000 BC on).

Another one on Steppe ancestry expansion, with a tentative distribution of “steppe ancestry” divided into that of Sredni Stog/Corded Ware origin vs. that of Repin/Yamna origin, a difference that has been known for quite some time already.

It is tentative because there hasn’t been any professional study or amateur attempt to date to differentiate both “steppe ancestries” in Yamna, and especially in Bell Beakers. So much for the call of professional geneticists since 2018 (see here and here) and archaeologists since 2017 (see e.g. here and here) to distinguish fine-scale population structure to be able to follow neighbouring populations which expanded with different archaeological (and thus ethnolinguistic) groups.

Tentative map of fine-scale population structure during steppe-related expansions (ca. 3500–2000 BC), including Repin–Yamna–Bell Beaker/Balkans and Sredni Stog–Corded Ware groups. Data based on published samples and pairwise comparisons tested to date. Notice that the potential admixture of expanding Repin/Early Yamna settlers in the North Pontic area with the late Sredni Stog population (and thus Sredni Stog-related ancestry in Yamna) has been omitted for simplicity purposes, assuming thus a homogeneous Yamna vs. Corded Ware ancestry.

I think both maps are especially important today, given the current Nordicist reactionary trends arguing (yet again) for an origin of Indo-Europeans in The North™, now based on the Fearsome Tisza River hypothesis, on cephalic index values, and a few pairwise comparisons – i.e. an absolutely no-nonsense approach to the Indo-European question (LOL). At least I get to relax and sit this year out just observing how other people bury themselves and their beloved “steppe ancestry=IE” under so many new pet theories…

NOTE. Not that there is anything wrong with a northern origin of North-West Indo-European from a linguistic point of view, as I commented recently – after all, a Corded Ware origin would roughly fit the linguistic guesstimates, unlike the proposed ancestral origins in Anatolia or India. The problem is that, like many other fringe theories, it is today just based on tradition, or (even worse) ethnic, political, or personal desires, and it doesn’t make sense when all findings from disciplines involved in the Indo-European and Uralic questions are combined.

Simple ancestry percentages in modern populations. Recent image by Iain Mathieson 2019 (min. 5.57). A simplistic “Steppe ancestry” defining Indo-European speakers…? Sure.

Within 20 or 30 years, when genetic genealogists (or amateur geneticists, or however you want to call them) ask why we had the opportunity since 2015 to sample as many Hungarian Yamnaya individuals as possible and we didn’t, when it is clear that the number of unscathed kurgans is diminishing every year (from an estimated 4,000 in the 20th century, of the original tens of thousands, to less than 1,500 today) the answer will not be “because this or that archaeologist or linguist was a dilettante or a charlatan‘, as they usually describe academics they dislike.

It will be precisely because the very same genetic genealogists – supposedly interested today in the origin of R1b-L151 and/or genetic marker associated with North-West Indo-Europeans – are obsessed with finding them anywhere else but for Hungary, and prefer to use their money and time to play with a few statistical tools within a biased framework of flawed assumptions and study designs, obtaining absurd results and accepting far-fetched interpretations of them, to be told exactly what they want to hear: be it the Franco-Cantabrian homeland, the Dutch or Moravian Beaker from CWC homeland, the Maykop homeland, or the Moon homeland.

Poetic justice this heritage destruction, whose indirect causes will remain written in Internet archives for everyone to see, as a good lesson for future generations.

Contrastive principal component analysis (cPCA) to explore patterns specific to a dataset

Interesting open access paper Exploring patterns enriched in a dataset with contrastive principal component analysis, by Abid, Zhang, Bagaria & Zou, Nature Communications (2018) 9:2134.

Abstract (emphasis mine):

Visualization and exploration of high-dimensional data is a ubiquitous challenge across disciplines. Widely used techniques such as principal component analysis (PCA) aim to identify dominant trends in one dataset. However, in many settings we have datasets collected under different conditions, e.g., a treatment and a control experiment, and we are interested in visualizing and exploring patterns that are specific to one dataset. This paper proposes a method, contrastive principal component analysis (cPCA), which identifies low-dimensional structures that are enriched in a dataset relative to comparison data. In a wide variety of experiments, we demonstrate that cPCA with a background dataset enables us to visualize dataset-specific patterns missed by PCA and other standard methods. We further provide a geometric interpretation of cPCA and strong mathematical guarantees. An implementation of cPCA is publicly available, and can be used for exploratory data analysis in many applications where PCA is currently used.

Schematic Overview of cPCA. To perform cPCA, compute the covariance matrices C X , C Y of the target and background datasets. The singular vectors of the weighted difference of the covariance matrices, C X  − α · C Y , are the directions returned by cPCA. As shown in the scatter plot on the right, PCA (on the target data) identifies the direction that has the highest variance in the target data, while cPCA identifies the direction that has a higher variance in the target data as compared to the background data. Projecting the target data onto the latter direction gives patterns unique to the target data and often reveals structure that is missed by PCA. Specifically, in this example, reducing the dimensionality of the target data by cPCA would reveal two distinct clusters

The Mexican example caught my attention:

Relationship between ancestral groups in Mexico

In previous examples, we have seen that cPCA allows the user to discover subclasses within a target dataset that are not labeled a priori. However, even when subclasses are known ahead of time, dimensionality reduction can be a useful way to visualize the relationship within groups. For example, PCA is often used to visualize the relationship between ethnic populations based on genetic variants, because projecting the genetic variants onto two dimensions often produces maps that offer striking visualizations of geographic and historic trends26,27. But again, PCA is limited to identifying the most dominant structure; when this represents universal or uninteresting variation, cPCA can be more effective at visualizing trends.

The dataset that we use for this example consists of single nucleotide polymorphisms (SNPs) from the genomes of individuals from five states in Mexico, collected in a previous study28. Mexican ancestry is challenging to analyze using PCA since the PCs usually do not reflect geographic origin within Mexico; instead, they reflect the proportion of European/Native American heritage of each Mexican individual, which dominates and obscures differences due to geographic origin within Mexico (see Fig. 4a). To overcome this problem, population geneticists manually prune SNPs, removing those known to derive from Europeans ancestry, before applying PCA. However, this procedure is of limited applicability since it requires knowing the origin of the SNPs and that the source of background variation to be very different from the variation of interest, which are often not the case.

Relationship between Mexican ancestry groups. a PCA applied to genetic data from individuals from 5 Mexican states does not reveal any visually discernible patterns in the embedded data. b cPCA applied to the same dataset reveals patterns in the data: individuals from the same state are clustered closer together in the cPCA embedding. c Furthermore, the distribution of the points reveals relationships between the groups that matches the geographic location of the different states: for example, individuals from geographically adjacent states are adjacent in the embedding. c Adapted from a map of Mexico that is originally the work of User:Allstrak at Wikipedia, published under a CC-BY-SA license, sourced from https://commons.wikimedia.org/wiki/File:Mexico_Map.svg

As an alternative, we use cPCA with a background dataset that consists of individuals from Mexico and from Europe. This background is dominated by Native American/European variation, allowing us to isolate the intra-Mexican variation in the target dataset. The results of applying cPCA are shown in Fig. 4b. We find that individuals from the same state in Mexico are embedded closer together. Furthermore, the two groups that are the most divergent are the Sonorans and the Mayans from Yucatan, which are also the most geographically distant within Mexico, while Mexicans from the other three states are close to each other, both geographically as well as in the embedding captured by cPCA (see Fig. 4c). See also Supplementary Fig. 6 for more details.

So, by using a background dataset, it discovers patterns in a single target dataset via dimensionality reduction, that standard dimensionality reduction techniques do not discover. Maybe useful for some prehistoric populations, too…

They have released a Python implementation of cPCA on GitHub: https://github.com/abidlabs/contrastive, including Python notebooks and datasets.

See also: