Mosaic maternal ancestry in the Great Lakes region of East Africa

Abstract

The Great Lakes lie within a region of East Africa with very high human genetic diversity, home of many ethno-linguistic groups usually assumed to be the product of a small number of major dispersals. However, our knowledge of these dispersals relies primarily on the inferences of historical linguistics and oral traditions, with attempts to match up the archaeological evidence where possible. This is an obvious area to which archaeogenetics can contribute, yet Uganda, at the heart of these developments, has not been studied for mitochondrial DNA (mtDNA) variation. Here, we compare mtDNA lineages at this putative genetic crossroads across 409 representatives of the major language groups: Bantu speakers and Eastern and Western Nilotic speakers. We show that Uganda harbours one of the highest mtDNA diversities within and between linguistic groups, with the various groups significantly differentiated from each other. Despite an inferred linguistic origin in South Sudan, the data from the two Nilotic-speaking groups point to a much more complex history, involving not only possible dispersals from Sudan and the Horn but also large-scale assimilation of autochthonous lineages within East Africa and even Uganda itself. The Eastern Nilotic group also carries signals characteristic of West-Central Africa, primarily due to Bantu influence, whereas a much stronger signal in the Western Nilotic group suggests direct West-Central African ancestry. Bantu speakers share lineages with both Nilotic groups, and also harbour East African lineages not found in Western Nilotic speakers, likely due to assimilating indigenous populations since arriving in the region ~3000 years ago.

Quantifying the legacy of the Chinese Neolithic on the maternal genetic heritage of Taiwan and Island Southeast Asia

“…Although the frequency of these lineages throughout the various aboriginal groups was never greater than 50 % (except in the Ami, with 56 %), it appears therefore that the admixture that generated the current maternal gene structure of Taiwanese tribes took place progressively after the “out-of-Taiwan” expansion. At the time the “out-of-Taiwan” dispersal occurred, the expanding population seems to have remained distinct, with a full South Chinese ancestry (similar to what also seems to have happened to a considerable degree in Central Europe (Haak et al 2010 ) and in the Great Lakes region during the Bantu expansion (Gomes et al 2015 ; Silva et al 2015 ). Whether the subsequent admixture with autochthonous groups (and, presumably, the assimilation of the language) was rapid or protracted we cannot assess from contemporary data alone, in the absence of ancient DNA evidence.…”

Mapping human dispersals into the Horn of Africa from Arabian Ice Age refugia using mitogenomes

“…More generally, the Horn of Africa is exceptional in harbouring very high mtDNA haplogroup diversity 4 , and populations in the Horn have significant non-autochthonous African ancestry across the genome 5 6 7 8 9 10 . Recent studies of complete human genomes have concluded that this 30–50% of non-African legacy in Cushitic- and Semitic-speaking populations is the result of admixture from Arabia beginning ~3,000 years ago (3 ka) 11 12 , at a time when common cultural features developed across the Horn and southern Arabia 13 , suggesting a link with the origin of the Ethiosemitic languages 14 .…”

Carriers of mitochondrial DNA macrohaplogroup L3 basic lineages migrated back to Africa from Asia around 70,000 years ago

“…The data also point to important sex-biased dispersals between populations. These evident gene replacements in Africa have been mainly attributed to recent geographic range expansions of pastoralist and agriculturalist populations from eastern and western Africa at the expense of the hunter-gatherers inhabitants of the Central Africa rainforest [108–110], eastern African forested areas around the Great Lakes [111–114], and the semi-desert open spaces of South Africa [115–118]. Under our hypothesis of an early return to Africa from Eurasia of basic mtDNA L3 and Y-chromosome E lineages, and their expansion around 70 kya first into East and later into West Africa, those lineage replacements must have begun very early.…”

“…The same “centre-of-gravity” argument was used by other authors to suggest a Central African origin [126]. It is worth mentioning that while ancestral southern African Khoesan-speaking population still maintain high frequencies of primitive L0d and k lineages [94, 106, 127, 128], and that in the hunter-gatherer populations of central-western Africa the L1c haplogroup is dominant [108, 109], L5 in eastern Africa has today only a marginal presence [114, 129], most probably due to its displacement produced by more recent waves of better adapted incomers. The presence of L5 in southern Africa and eastern Mbuti pygmies [70, 109, 118, 127] is the result of later migrations.…”

“…Its first split was haplogroup L6 that nowadays is a rare eastern lineage with a deep founder age (about 100 kya) but a rather recent expansion (about 25 kya). It has been found at frequencies below 1% in Egyptians [131], Somalis [132], Kenyan [133], and eastern Nilotes from Uganda [114]. Mean frequency rise in Ethiopia (3.15 ± 1.15 %) with a maximum (15.8%) in Ongota, an extinguishing linguistic isolate of uncertain adscription [129].…”