Archaeogenetics of two subcontinents:the transition to Metal Ages in South Asia and Southwest Europe

The transition to the Metal Ages led to profound economic and social changes in Eurasian populations, as seen from the archaeological record. The spread of the Indo-European language family, from a putative origin in the Eurasian steppe, across Europe and Central and South Asia accompanied population movements and interactions during the Chalcolithic and Bronze Age. The goal of this project is to study the demographic dynamics of human populations during the transition to Metal Ages across Eurasia, by focusing on the two extremes of the Indo-European distribution: South Asia and Iberia. Additional results from a pilot study on the Medieval period in Iberia are also presented.

The comparison of uniparental markers in South Asia shows that there is a strong sex bias in the ancestry of present-day South Asian populations: while the maternal variation is mostly autochthonous (~70–90%), dating to the first settlement of the region, the paternal gene pool reflects more recent ancestry from Southwest Asia and Central Asia (~50–90%), mostly in the form of R1a branches dating to the Bronze Age.

However, despite the male-biased nature of Bronze Age movements, minor signals can also be found in the maternal gene pool. Haplogroup H is the most common mitochondrial lineage in present-day European populations (up to 45%), and is also present at considerable frequencies in other regions, such as Central and South Asia. Haplogroup H was present across continental Europe since the Neolithic, and it is possible to distinguish Late Neolithic/Bronze Age signals in some H lineages: H2 and H13 were incorporated in movements across the Steppe into Europe, in the west (mainly as H2a and H13a1a), and across Central Asia and into South Asia, in the east, mainly in the form of H2b, although a minor signal is also visible for H13a1a and in H1b. In this thesis I show how a detailed phylogeographic approach, combining both modern and ancient variation can provide additional clues into population movements, even in the case of strong male bias. An increase in Steppe-related/Central-European ancestry in Iberia in the Bronze Age coincides with a large influx of Central-European Y-chromosome R1b lineages, but the same signal is not observed in the maternal gene pool. Changes in maternal variation in Iberia are only observed later, in the Iron Age, with an increase in the frequency of haplogroup H in general (and specifically H1), coinciding with a further increase in Central-European ancestry.

The Medieval period in Iberia is characterised by a substantial arrival of people from North Africa during the Islamic period, but their potential contribution to the Iberian gene pool was largely erased by several post-Medieval events. Here, I present a pilot study on aDNA from Medieval Iberia, with particular focus on the genome of an 11th century man buried in an Islamic necropolis in Segorbe, Spain (mean average genomic coverage: 0.065x). Uniparental lineages (mtDNA U6a1a1; Y-chromosome E1b1b1b1) indicate North African ancestry. However, at autosomal level he displays both North African and European-like ancestries. Formal tests of admixture indicate that he was most likely the result of admixture between two populations, a North African and a Spanish population (although the exact populations are difficult to pinpoint, due to possible increased Levantine ancestry in one of the sources), fitting historical accounts of intermarriages during this period and recently published aDNA evidence. In addition I present two mtDNA sequences (haplogroups H5a1 and V), the former of which represents the first example of Medieval genetic data from Portugal. These two sequences fall in mostly northwest European branches and strongly contrast with the U6a haplotype retrieved from the individual found in the Islamic cemetery.

Finally, I present preliminary analyses on a low coverage (0.016-0.871x) and SNP-capture dataset of 21 individuals from (Late) Neolithic/Chalcolithic/Bronze Age sites from Portugal and eastern and southern Spain. In agreement to previous findings for this period in Iberia, all Late Neolithic/Chalcolithic males carried either an I or a G2 Y-chromosome lineage, while the maternal gene pool was much more diverse. The Bronze Age individual from Cova L’Iguala (female) carried an mtDNA H3 haplotype, which is to our knowledge the first H3 reported for this period in Iberia, and is the only individual displaying Steppe ancestry at significant proportions. However, due to low coverage (0.03x) D-statistics results for this sample are not significant. Instead, this individual shares the highest drift with Iberian Middle/Late Neolithic, British Neolithic and Iberian Late Neolithic/Chalcolithic groups, similarly to the majority of the samples analysed in this chapter. I detect phenotypic diversity in Late Neolithic/Chalcolithic Portugal, as well as the presence of both ancestral and derived variants in SNPs related to immunity and vitamin D metabolism.