Neolithic farmers went (also) to West Africa

a) PCA, b) f3 analyses of admixture and c) admixture map of several African populations. Gallego Llorente et al. 2015 ©2015 American Association for the Advancement of Science. All Rights Reserved.


The main results of the paper, and specifically, the one I highlighted in the title of this article, have been debunked. Apparently a bioinformatics mistake was responsible for some of the most striking findings in Gallego-Llorente et al. 2015. See this article and the published erratum.


The Neolithic entered to Europe some 7000 years ago from the Near East. Not only the adoption of farming and herding, but also a radical cultural and behavioural change on how humans faced life came with them. The Neolithic Transition is one of the most interesting and more studied milestones in human evolution. The disjunctive on whether the process of spread of agriculture from Anatolia all the way to Iberia implied groups of resident hunter-gatherers learning farming activities and then teaching to others (cultural diffusion) versus the hypothesis of population replacement (demic diffusion) has inspired a huge load of research (see here or here for examples).

With the recent flourishing of ancient genomic techniques evidence overwhelmingly leans towards the demic diffusion hypothesis (see here). Earlier this year, Wolfgang Haak and his colleagues published more than a hundred ancient genomes from central Europe. While in Europe and Asia there is a growing number of acceptable quality ancient genomes to study these population movements, other areas of the world are not so lucky. The climatic conditions of warm and humid regions are more aggressive with human remains, impeding its conservation throughout millennia and proving very challenging to retrieve any DNA, although efforts are being done in order to change this (see here and here). As consequence, the number of ancient genomes available in very important areas to explain human history such as those surrounding the equator (Central Africa, Central America and South East Asia) is worrisomely close to zero.

Today, Marcos Gallego Llorente along with his colleagues from the Zoology Department in Cambridge, present in a nice paper the results of the sequencing of a 4500 years old male individual nicknamed ‘Mota’ found in Ethiopia. Mota was sequenced using whole-genome sequencing averaging a coverage of 12.5x throughout the genome and the authors applied to it the common genomic tools to discover its genetic affinities with modern populations and ancient genomes. Some results are remarkable:

As you can see in the Figure above, in model-free analyses -PCA and f3 statistics- Mota shows closer genetic affinities with the Ari groups, modern southern Ethiopians. The authors claim that this is may indicate population continuity between Mota and Ari. Although I think that model-free based methods are dramatically uninformative when addressing population continuity issues, an ancestral common origin for both Ari groups (Cultivators and Blacksmiths) fits with the results of an interesting recent paper from Lucy Van Drop and colleagues.

A previous work from Joseph Pickrell and colleagues had estimated ancient Eurasian backflow into all Khoisan populations dating ~3000 years before present. Because Mota dating was previous to that, they authors mixed it along with another West Eurasian population to model the most probably origin of that backflow. The results are surprising. European Neolithic early farmers from the Linearbandkeramik Culture (LBK-Stuttgart) and modern Sardinians show the closest affinities to the Ethiopian modelled that way, suggesting that:

The backflow [to Africa] came from the same genetic source that fuelled the Neolithic expansion into Europe from the Near East/Anatolia, before recent historic evens changed the genetic makeup of populations living in that region.

Interestingly, the authors estimated that all African populations have higher Eurasian ancestry acquired that way than previously estimated (3–4% more), and it reaches all the way to West and Southern Africa. This has great implications. West African populations such as Mbuti or Yoruba are usually regarded as Eurasian-ancestry free and therefore used as African reference in many studies. The results from Mota analyses indicated, however, that they have 6 and 7% of Eurasian ancestry, respectively, coming from the backflow. Inferences from many studies based in these reference populations may need to be revised, as Mota seems now the only unadmixed reference of African ancestry.

Mota genome was also screened in search of phenotypic markers such as the ones for eye, skin color and lactase persistence. Found genotypes indicated that Mota had dark skin and eye and was intolerant to lactase, not surprising given that it is Eurasian-ancestry free. Finally, the fact that it already had the selected alleles associated to adaptation to high altitude that are found in today’s Ethiopians may add some support to the author’s continuity hypothesis.

Altogether I consider this a very interesting, well explained, and easy to read, paper. I am looking forward to see what ancient genomes from other African locations have to say about the origins of modern human populations.