Bayesian estimation of partial population continuity using ancient DNA and spatially explicit simulations

Abstract

The retrieval of ancient DNA from osteological material provides direct evidence of human genetic diversity in the past. Ancient DNA samples are often used to investigate whether there was population continuity in the settlement history of an area. Methods based on the serial coalescent algorithm have been developed to test whether the population continuity hypothesis can be statistically rejected by analysing DNA samples from the same region but of different ages. Rejection of this hypothesis is indicative of a large genetic shift, possibly due to immigration occurring between two sampling times. However, this approach is only able to reject a model of full continuity model (a total absence of genetic input from outside), but admixture between local and immigrant populations may lead to partial continuity. We have recently developed a method to test for population continuity that explicitly considers the spatial and temporal dynamics of populations. Here, we extended this approach to estimate the proportion of genetic continuity between two populations, using ancient genetic samples. We applied our original approach to the question of the Neolithic transition in Central Europe. Our results confirmed the rejection of full continuity, but our approach represents an important step forward by estimating the relative contribution of immigrant farmers and of local hunter‐gatherers to the final Central European Neolithic genetic pool. Furthermore, we show that a substantial proportion of genes brought by the farmers in this region were assimilated from other hunter‐gatherer populations along the way from Anatolia, which was not detectable by previous continuity tests. Our approach is also able to jointly estimate demographic parameters, as we show here by finding both low density and low migration rate for pre‐Neolithic hunter‐gatherers. It provides a useful tool for the analysis of the numerous ancient DNA data sets that are currently being produced for many different species.

Simulated patterns of mitochondrial diversity are consistent with partial population turnover in Bronze Age Central Europe

“…We did not explore γ > 0.05 further because it always led to a final genetic contribution of Yamnaya equal to 0% as admixture was in this situation high enough for the Yamnaya genes to be fully “diluted” in the farming population before reaching Central Europe (Figure 3 ). This is a well‐known process in the context of the Neolithic transition (Chikhi et al, 2002 ; Silva et al, 2018 ), which has also been described to occur commonly in the situation of biological invasions with admixture (Quilodran et al, 2020 ).…”

SPLATCHE3: simulation of serial genetic data under spatially explicit evolutionary scenarios including long-distance dispersal

“…However, it is possible to simulate thousands or tens of thousands of independent or partly linked loci and use them as a proxy for full genome data through the computation of summary statistics (e.g. Currat and Excoffier, 2011; Silva et al , 2018). Running time is mostly affected by the number of simulated generations, number of demes and number of loci.…”

“…SPLATCHE3 can now simulate genetic data for serial samples through the implementation of the serial coalescent algorithm (Rodrigo and Felsenstein, 1999). This capability of SPLATCHE3 may be used to study aDNA and has already been used in two studies investigating population continuity through time (Silva et al , 2017, 2018). If one wants to add aDNA post-mortem features, scripts (e.g.…”

Ancient mitochondrial diversity reveals population homogeneity in Neolithic Greece and identifies population dynamics along the Danubian expansion axis

“…To investigate population continuity in northern Greece and the relationship between Neolithic farmers along the Danubian expansion route from the northern Aegean to the Balkans and central Europe, we adapted the spatially explicit simulation framework initially designed by Currat and Excoffier 70 and later improved by Silva et al 71 using a modified version of the program SPLATCHE2 72 . This framework allows the simulation of mitochondrial lineages at different points in time and space under alternative scenarios of population dynamics.…”

“…Each deme has a carrying capacity ( K HG ) of 100 effective haploid females, which corresponds to 200 individuals (males and females) per cell and a human density of 200*3/10,000 = 0.06 individuals/km 2 73 , 74 , assuming that the census size was three times the effective population size 75 . The migration rate ( m HG ) and growth rate ( r HG ) were set to 0.15 and 0.2, respectively, to achieve a colonization of Europe in approximately 500 generations, following Silva et al 71 (Table 1 ). The m HG represents the proportion of individuals in each deme emigrating to neighboring demes at each generation and the r HG represent the intrinsic rate of population growth per generation.…”