Population replacements and founder effects in humans

TREE has an article in press looking at the impact of founder events on subsequent patterns of genetic diversity. The main argument of the paper, co-authored by Waters, Fraser & Hewitt (Founder takes all: density-dependent processes structure biodiversity), is that density-dependent effects associated with primary colonization/replacement events play a major role in shaping patterns of population variation. Founder effects lead to clusters of differently homogenous populations depending on which genetic variants arrived first and had initial success.

When a species encounters new opportunities, such as through colonization of new islands, colonization of recently deglaciated areas, or the extinction of a sister lineage, density-dependent processes often become particularly important and can leave long-lasting genetic signatures [5]. Although several different terms have been proposed to explain such density-dependent phenomena (e.g., leading edge, high-density blocking, and gene surfing; Figure 1), we contend that these are all largely similar and interlinked. Despite their importance for studies of gene spread, phylogeography, competitive exclusion, and biogeography, the potential biological and biogeographic restrictions imposed by density-dependent processes have often been overlooked in the scientific literature. Such an oversight may result from the focus of many researchers on ‘what has happened?’ rather than ‘what hasn’t happened?’ in ecology and evolution.

What caught my attention is that one of the examples they use in support of their argument is the “Out of Africa” expansion of modern humans. I find this a strange example for the case they are making, not because it undermines it exactly, but because it points out the tremendous complexity that can go into replacement/founder events. The authors’ argument is that initial colonizers leave lasting genetic signals because, via effects correlated with density, they are able to exclude other potential genetic variants. Applied to humans, there are a number of potential expectations you might draw, none of which provide an obvious “founder takes all” scenario.

For example, you might argue that Neandertals are the first (kind of, sort of) human population to occupy Europe, either expanding into the continent in the Mid-Pleistocene or developing out of in situ populations within Europe during that time, and subsequently waxing and waning in frequency and range throughout glacial-interglacial cycles. In this case, you might expect living Europeans to preserve a predominately Neandertal genetic signal. But obviously we do not. Most Eurasians preserve a signal of this ancestry, just not a very large one (Yang et al., 2012).

So a second way of viewing it would be that while Neandertals were the first, they were taken advantage of and more or less completely replaced by early modern humans in the 40,000-25,000 year range. This replacement was extensive enough that we might view these new human occupants as the founding population, and genetically similar to contemporary European populations. Only they are not, at least not uniformly so (Skoglund et al. 2012).

Instead, contemporary European populations are most similar to Neolithic agricultural populations that expanded out of the Near East within the past 10,000 years (Haak et al. 2010, Balaresque et al. 2010, Fu et al. 2012). It is hard to justify this population as a “founding” population in Europe in a evolutionary/demographic sense.

These kinds of population replacement events and the demographic/genetic effects the authors of this paper describe are certainly important. But in the case of human population replacements, the reality seems more complex and more case-specific than the argument provided. I would argue that the fossil, archaeological and genetic record suggest that human population replacements (and correlated collapses) are relatively common, but the pattern of continuity and discontinuity is dependent on the exact mechanisms of collapse and pattern of replacement. The demographic pattern of glacial and interglacial Neandertals, Mesolithic hunter-gatherers, and Neolithic farmers have significant differences. Furthermore, the culturally-mediated technological modifications of the expanding populations, in each case, are also quite different, setting up different possibilities for the extent of replacement versus admixture processes.

The history of human populations is dynamic and complex. There is value in thinking about general models that allow the details of specific prehistoric population events to be teased apart, but I doubt the utility of generalized explanations of human prehistory.

*****

1. Waters, Jonathan M., Fraser, Ceridwen I., Hewitt, Godfrey M. (2012) Founder takes all: density-dependent processes structure biodiversity, Trends in Ecology & Evolution. doi: 10.1016/j.tree.2012.08.024

2. Yang, M. A., A.-S. Malaspinas, et al. (2012). “Ancient Structure in Africa Unlikely to Explain Neanderthal and Non-African Genetic Similarity.” Molecular Biology and Evolution 29(10): 2987-2995. doi:10.1093/molbev/mss117

3. Pontus Skoglund,Helena Malmström, Maanasa Raghavan, Jan Storå, Per Hall, Eske Willerslev, M. Thomas P. Gilbert, Anders Götherström, Mattias Jakobsson (2012). Origins and Genetic Legacy of Neolithic Farmers and Hunter-Gatherers in Europe. Science 336, 466. DOI: 10.1126/science.1216304.

4. Haak, W., O. Balanovsky, et al. (2010). “Ancient DNA from European Early Neolithic Farmers Reveals Their Near Eastern Affinities.” PLoS Biol 8(11): e1000536. doi:10.1371/journal.pbio.1000536

5. Fu, Q., P. Rudan, et al. (2012). “Complete Mitochondrial Genomes Reveal Neolithic Expansion into Europe.” PLoS ONE 7(3): e32473. doi:10.1371/journal.pone.0032473

6. Balaresque, P., G. R. Bowden, et al. (2010). “A Predominantly Neolithic Origin for European Paternal Lineages.” PLoS Biol 8(1): e1000285. doi:10.1371/journal.pbio.1000285