There's been a lot of talk lately about the Copper Age being a key period for the emergence of the modern European gene pool, mainly thanks to the expansions of pan-European cultures like the Bell Beaker and Corded Ware. I agree with this line of thinking, but let's see what details we can extrapolate from a couple of Principal Component Analyses (PCAs) based on a wide variety of modern and ancient mtDNA samples available online. The first PCA is limited to samples of predominantly West Eurasian origin, while the second also features Central and East Asian samples. Please note, most of the work was outsourced to Eurogenes project member PL16. The dataset is available here. It includes some samples that haven't yet been formally published (ie. peer reviewed).
Key: AK - Andronovo, Karakol and Karasuk cultures; AS - Medieval Anglo-Saxon; Adyg - Adygei; Assyr - Assyrian; Azer - Azeri; BB - Bell Beaker Culture from Germany and Denmark; BOO - Bolshoy Oleni Ostrov Mesolithic site; BY - Belorussian; Bashkir - Bashkir; Bo - Bologna; C-I - Central Italy; CW - Corded Ware Culture; CZ - Czech; DD - Dnieper-Donets Culture; DE - Germany; EUP - European Upper Paleolithic; FB - Funnelbeaker Culture mixed Farmer/Forager sample; IAD - Iron-Age Danish; IES - Iron Age Eastern Spanish; INCS - Iberian Neolithic; IR - Ireland; IS - Iceland; Kazakhs - Kazakhs; Kirghiz - Kirghiz; MBC - Medieval Basque Country; MH - Medieval Hungarian; MP - Medieval Poland; M - Minoan; Mongols - Mongols; NEI - Northeast Italy; NRU - North Russian; NWI - Northwest Italy; Nogays - Nogays; Nur - Nuragic Sardinian; Orkn - Orkney; PCA - Portuguese Copper Age; PW - Przeworsk and Wielbark Cultures; PitW - Pitted Ware Culture; PolK - Poland Kashuby; PolP - Poland Podhale; RUCos - Cossack; S-I - South Italy; SRU - South Russian; Saam - Saami; ScSb - Scytho-Siberian; Sc - Don Scythian; Scan - Scandinavia; Scot - Scotland; Sic - Sicily; Srd - Sardinia; TRI - Tripolye Culture; T - Treilles Neolithic site; Turk - Turkish; UA - Ukraine; UOO - Uznyi Oleni Ostrov Mesolithic site; UTr - Ust-Tartas Copper Age site; Udmurt - Udmurt; Uigurs - Uigurs; Une - Unetice or Aunjetitz Culture; Urn - Urnfield; VD - Viking Danish; WCI - West-Central Italy (inc. Tuscany); Xion - Xiongnu.The main thing to keep in mind about PCA is that it reduces the data so that (hopefully) only the most important information is retained. So it's interesting to note that the most significant variable in both analyses is the frequency of mtDNA H (as per the Loadings plots). Moreover, the PCAs show that the frequency of mtDNA H increased rapidly across much of Europe just after the Copper Age, which is very much in line with what I mentioned in the intro (for more on that, see here and here). Interestingly, the data suggest this shift might have originated in Copper Age Portugal. That's because the Portuguese Copper Age sample (PCA) is the oldest group with a modern European or greater than frequency of mtDNA H (>45%). Indeed, its mtDNA H frequency of 74% is the highest in the entire dataset. Another really interesting thing is the placement of almost all the modern European samples near the combined Copper Age Bell Beaker sample from Germany and Denmark (BB) - in other words, within the so called Central European Bell Beaker cluster. Based on its position this cluster could well be described as a mix of the nearby clusters - Atlanto-Mediterranean, Mixed Farmer/Hunter-Gatherer, and Anatolian. This result correlates well with latest archeological and anthropological findings, which suggest the Bell Beaker folk originated in Copper Age Portugal and subsequently diffused across most of Western and Central Europe. Please note, the clusters aren't simply based on the PCA results, but also on archeological and geographic affinities. So admittedly it's an ad hoc approach, but I think it works well in this case. Here's a brief rundown of what most of the clusters represent.
Atlanto-Mediterranean: This cluster is well defined by the first PC on both plots, and includes samples from Copper Age Portugal (PCA), Bronze Age Nuragic Sardinia (Nur) and Medieval Basque Country (MBC). Perhaps this is also where the mtDNA of the earliest Portuguese Bell Beaker folk would land? Central European Bell Beaker: Based on this cluster it looks like we're all Central European Bell Beakers (BB), in varying degrees anyway. Mixed Farmer/Hunter-Gatherer: I'd say this cluster shows what mtDNA haplogroup frequencies looked like across much of Europe just before the Bell Beaker expansion. It includes Corded Ware (CW) and Unetice (Une) Culture samples from Copper and Bronze Age Germany, respectively. These samples belong to archeological cultures often classified as Indo-European, and appear much more eastern than the Copper Age Central European Bell Beaker sample and most modern Europeans. However, they're fairly similar to the earlier farmer/forager Funnelbeaker (FB) sample from North Germany and Southern Scandinavia. So it's not possible to say with any great certainty at the moment where they originated. My guess is the forest steppe just east of Germany. Anatolian: This cluster falls between two European clusters along the first PC, which is probably an unexpected result. There are three main reasons for this: a) the data doesn't distinguish between subclades of mtDNA H, which often show markedly different frequencies between Europe and West Asia; b) many of the prehistoric European samples carry lower levels of mtDNA H than modern West Asian samples; c) Europeans are in large-part the descendants of Neolithic migrants from the Near East, while some West Asians, like the Azeris, seem to have significant European admixture. Steppe & Forest Steppe: The Dnieper-Donets sample (DD) from the late Neolithic Ukrainian steppe falls within this cluster, which mostly includes modern populations of mixed West and East Eurasian origin. Now, the Dnieper-Donets Culture is sometimes classified as part of the Proto-Indo-European (PIE) archeological sphere. If so, then it seems the Proto-Indo-Europeans did not have a significant impact on the mtDNA of modern Europeans. Hunter-Gatherer: The Hunter-Gatherer clusters on both PCAs are similar and mostly include prehistoric samples with extreme levels of mtDNA haplogroups U4, U5 and/or U. However, the European Hunter-Gatherer cluster on the second plot is limited to samples with low or non-existent frequencies of East Eurasian-specific mtDNA haplogroups, like C.Some of these clusters show a close correlation with those from my ADMIXTURE analyses based on genome-wide data from modern samples. For instance, the Mediterranean cluster from the K=12 run (see here) peaks in the West Mediterranean region and looks a lot like the modern autosomal equivalent of the Atlanto-Mediterranean cluster described above. Indeed, as per the Fst (genetic) distance table below, the Mediterranean cluster (pop1) is the most distinct European-specific cluster. The North Atlantic (pop7) and Baltic (pop10) clusters are actually closer to the Caucasus or West Asian cluster (pop6) than to the Mediterranean cluster. As far as I can recall, this pattern was repeated in all the other relevant ADMIXTURE tests I've run.
But all of this still leaves us with the question of how this genetically peculiar population came to be in Western Europe in the first place. It probably wasn't indigenous to Iberia, because a couple of Mesolithic hunter-gatherer skeletons from northwestern Spain were recently analyzed with tens of thousands of genome-wide SNPs, and came out more Northern European than Mediterranean (see here). These indigenous Iberians also belonged to the typically hunter-gatherer mtDNA haplogroup U5. Perhaps it arrived in Europe during the Neolithic? However, on the PCAs above the Neolithic Iberian sample (INCS) is clearly not of the same Atlanto-Mediterranean origin. It's actually placed with the Corded Ware (CW) and Unetice (Une) samples. As I recently commented after seeing the Minoan aDNA from Hughey et al. (see here), perhaps the ancestors of the Bell Beakers originated in West Asia, and were a very specific subset of the genetic diversity found in that region during the late Neolithic? But after reviewing the ancient mtDNA data here, I'm not so sure anymore. To be honest, the lost civilization of Atlantis is now starting to look like a really good option. Update 15/06/2013: Below is a new PCA with additional samples and hopefully more detail. The updated dataset can be downloaded here.
It's interesting to note the close proximity of the Unetice (Une) and Corded Ware (CW) cultures to modern eastern Belorussians (BE), Komi Zyrians (Komi-Z) and Chuvashs. Indeed, the Unetice, Corded Ware and eastern Belorussian samples land right next to each other on PCAs of all possible combinations of PC1 to PC5, but the Komi and Chuvashs clearly pull away from the ancient Central Europeans across some of the PCs.
See also... Bronze Age Germans came from...eastern Belarus? References... New Vistas on the Distant Past: Ancient Western Eurasian DNA Balanovsky Oleg P., Variability of gene pool in space and time: Data synthesis, genogeography, mitochondrial DNA and Y-chromosome. Dissertation for the degree of Doctor of Science. 2012 Bekada A, Fregel R, Cabrera VM, Larruga JM, Pestano J, et al. (2013) Introducing the Algerian Mitochondrial DNA and Y-Chromosome Profiles into the North African Landscape. PLoS ONE 8(2): e56775. doi:10.1371/journal.pone.0056775 Bermisheva et al., Phylogeographic Analysis of Mitochondrial DNA in the Nogays: A Strong Mixture of Maternal Lineages from Eastern and Western Eurasia, Molecular Biology, July 2004, Volume 38, Issue 4, pp 516-523, doi:10.1023/B:MBIL.0000037003.28999.45 Bermisheva et al., Diversity of Mitochondrial DNA Haplogroups in Ethnic Populations of the Volga–Ural Region, Molecular Biology, Vol. 36, No. 6, 2002, pp. 802–812. PMID: 12500536 Boattini A, Martinez-Cruz B, Sarno S, Harmant C, Useli A, et al. (2013) Uniparental Markers in Italy Reveal a Sex-Biased Genetic Structure and Different Historical Strata. PLoS ONE 8(5): e65441. doi:10.1371/journal.pone.0065441 Brotherton et al., Neolithic mitochondrial haplogroup H genomes and the genetic origins of Europeans, Nature Communications 4, Article number: 1764, Published 23 April 2013, doi:10.1038/ncomms2656 Hughey et al., A European population in Minoan Bronze Age Crete, Nature Communications 4, Article number: 1861, doi:10.1038/ncomms2871, Published 14 May 2013 Helgason et al., mtDNA and the Islands of the North Atlantic: Estimating the Proportions of Norse and Gaelic Ancestry, Am. J. Hum. Genet. 68:723–737, 2001, PMCID: PMC1274484 Kushniarevich A, Sivitskaya L, Danilenko N, Novogrodskii T, Tsybovsky I, et al. (2013) Uniparental Genetic Heritage of Belarusians: Encounter of Rare Middle Eastern Matrilineages with a Central European Mitochondrial DNA Pool. PLoS ONE 8(6): e66499. doi:10.1371/journal.pone.0066499