Influence of the last glacial period on the genetic diversity of modern humans

Zusammenfassung

During the last decades, multiple studies focused on the evolutionary history of our species, however this topic still remains without a clear and consensual answer. A key environmental factor that might have deeply influenced the genetic patterns of current modern humans is the last glacial period (LGP). It is known that this period promoted population range contractions, refugia isolation and posterior re-expansions (e.g., in the mainland Asia), but it could also allow a rapid arrival to regions such as the Americas and Southeast Asian islands due to the induced sea level lowstand. Additionally, other factors (e.g., posterior expansions, population admixture and different migratory routes and patterns) may have also influenced the spatial genetic patterns of current populations. These spatiotemporal dynamics of populations of our species is described in Chapter 1. The main goal of this PhD Thesis is to evaluate formally the influence of the LGP on modern humans and understand its role on the spatial genetic diversity of current populations. Analyses of the genetic variation of current populations show spatial genetic gradients worldwide, however their causes and relationships with past environmental processes (such as the LGP) are unclear. In this concern, in Chapter 2 and Chapter 3 I investigated the influence of different evolutionary scenarios on the spatial genetic gradients of American and Asian populations, respectively, to evaluate which evolutionary process better explains the currently observed genetic patterns. Using spatially-explicit computer simulations, I evaluated the influence of (i) admixture between different expansion waves of modern humans, (ii) the presence of ice sheets during the last glacial maximum (LGM) and (iii) long-distance dispersal (LDD) events, on current genetic gradients. Next, I compared the genetic gradients derived from data simulated under each evolutionary scenario with genetic gradients derived from real genetic data. Regarding the American continent, the results showed a gradient with a northwest-southeast (NW-SE) orientation, common to all the evaluated scenarios, and in agreement with genetic gradients obtained from real data. This gradient seems to be a consequence of isolation-by-distance along the long distance of this continent. I also investigated the genetic gradient in North and South America separately, which curiously showed an orientation orthogonal to the direction of the expansion, and that I propose to be caused by an allele surfing process. Importantly, this process in North America could be derived from the melting of the ice sheets in the end of the LGM. Overall, these findings show a global genetic gradient in the American continent but also distinct genetic gradients among local regions of the continent. Concerning the Asian continent, I found that the spatial genetic gradients were sensible to the applied evolutionary processes, especially the range contraction induced by the LGM and the number of Neolithic expansion waves. In particular, I obtained a genetic gradient with an east-west (E-W) orientation (i.e., similar to that obtained from real data) caused by two Neolithic expansions (from Middle East and from East Asia) or by considering a Paleolithic population suffering a range contraction induced by the LGM. Overall, the results suggest that the genetic gradients of Paleolithic populations were severely influenced by changes of the living range induced by the LGP. We interpreted the observed genetic gradients as a consequence of both allele surfing in range expansions and isolation-by-distance over the long E-W geographic distance of this continent. Analogously to the Americas, the sea level lowstand induced by the LGP may have facilitated the colonization of the Southeast Asian islands. In particular, during the LGP currently submersed lands were exposed connecting distant islands through land bridges. Still, little is known about the influence of the lowering sea level on the genetic diversity of current populations from Southeast Asia (SEA). Thus, in Chapter 4 I evaluated the fitting of alternative scenarios of the settlement of our species in the Southeast Asian islands with current genetic data. In particular, I applied approximate Bayesian computation (ABC), based on extensive spatially-explicit computer simulations, to evaluate the fitting of mitochondrial DNA (mtDNA) data from a variety of current local populations with alternative evolutionary scenarios that consider and ignore the LGP. Also, considering that some islands remained isolated, maritime migration could have played a key role in the region, hence I also evaluated scenarios presenting (and ignoring) LDD events. The results show that both the LGP and migration through LDD favoured a rapid expansion over SEA and should be taken into consideration to explain the currently observed genetic diversity in these populations. Also, the results showed that the temporary lands provided additional resources and gene flow corridors that favoured genetic diversity. Overall, the LGP may have promoted an increase of genetic diversity of current human populations from SEA. In the final Chapter of this Thesis, I focused on the routes of first successful out-of-Africa migration of anatomically modern humans (AMHs). Two routes from eastern Africa were proposed (along the Nile Valley until the Levant and/or crossing the Red Sea through the Bab el Mandab Strait until the Arabian Peninsula), yet they are still very contentious. In Chapter 5 I evaluated the fitting of two alternative evolutionary scenarios differing on the migratory routes with genetic data from current human population. In particular, I applied ABC based on extensive spatially-explicit genetic simulations, to evaluate the fitting of both migration routes (along Nile Valley or crossing the Red Sea) with genome-wide single nucleotide polymorphisms (SNPs) data that belongs to populations located around the geographic area of origin of the out-of-Africa migration (eastern Africa, Middle East and South Asia). The results suggest that the first successful out-of-Africa migration of AMHs could have occurred first by crossing the Red Sea and later along the Nile Valley, but a higher number of migrants crossed the Red Sea. This may be a consequence of the sea level lowstand during the LGP that approached the east and west shores of the Red Sea reducing the crossing distance from Africa to Asia. Also, the results show that the glacial refugia in Arabian Peninsula provided resources to temporarily isolated populations and prevented depopulation of the peninsula. Overall, the studies conducted along this Thesis formally show that the environmental fluctuations during the LGP played a key role in the expansion of AMHs and influenced the genetic diversity of current worldwide human populations. Specifically, the LGP eased the expansion of AMHs and contributed to the increase of genetic diversity in some regions, while in other regions it delayed the settlement of AMHs and reduced genetic diversity.