Archaeological evidence shows that, in the long run, Neolitization (the transition from foraging to food production) was associated with demographic growth. We used two methods (patterns of linkage disequilibrium from whole-genome SNPs and MSMC estimates on genomes) to reconstruct the demographic profiles for respectively 64 and 24 modern-day populations with contrasting lifestyles across the Old World (sub-Saharan Africa, south-eastern Asia, Siberia). Surprisingly, in all regions, food producers had larger effective population sizes (Ne) than foragers already 20 k years ago, well before the Neolithic revolution. As expected, this difference further increased ~12–10 k years ago, around or just before the onset of food production. Using paleoclimate reconstructions, we show that the early difference in Ne cannot be explained by food producers inhabiting more favorable regions. A number of mechanisms, including ancestral differences in census size, sedentism, exploitation of the natural resources, social stratification or connectivity between groups, might have led to the early differences in Ne detected in our analyses. Irrespective of the specific mechanisms involved, our results provide further evidence that long term cultural differences among populations of Palaeolithic hunter-gatherers are likely to have played an important role in the later Neolithization process.
The advent of food production marked a shift in human history entailing important changes in technology (e.g. mills, plant and animal domestication, use of ceramics), economy (e.g. accumulation of goods) and society (e.g. sedentism). Several lines of evidence point to this revolution leading to an increase in population density1,2,3,4.
Such growth is expected to have left a signature in the genomes, through a change in Ne, the effective population size. Indeed, gene genealogies of expanding populations should show an excess of singletons and private alleles when compared with those of stationary populations5. Because recombination brings together in the same chromosome DNA tracts with different genealogies, analysis of non-recombining DNA regions (mostly in mitochondrial DNA and in the Y chromosome) is often the simplest way to investigate past changes in population size.
Genetic markers of modern day populations with different lifestyles have been compared since the late '90s, with a number of studies finding significant demographic differences between food producers and hunter-gatherers6,7,8,9,10. More recently, several studies have used mtDNA to date the beginning of effective population growth in food producers, and, unexpectedly, the increase in Ne was inferred to have started long before the Neolithic transition11,12,13,14. However, dating based on mtDNA can be challenging15, and evidence from other markers is limited. Work on the Y chromosome16,17 and a few autosomal loci10,18,19 lend some support to the notion of demographic changes predating food production, but their coverage in terms of populations is limited. The most comprehensive study to date on this topic in term of both markers and geographic coverage18, only included 20 autosomal loci for 16 populations, and these mostly covered Africa (n = 10), whilst the remaining 6 Eurasian populations included several highly urbanized ones (Danes, Han Chinese and Japanese) for which the more recent history could have strongly biased this kind of estimates.
Beside the suggested cultural differences (subsistence strategies, sedentism, etc.), a hypothesis that should also be taken into account is that the observed discrepancies could be the result of a differential geographic distribution of the resources20,21. A simple scenario might be envisaged where populations that lived in climatically more favorable areas started growing well before the advent of food production. The same favorable climate would also make the adoption of food production more likely, as it would allow for the growth of crops and a sedentary lifestyle. By contrast, populations in harsher environments might have retained hunting and gathering (and probably some level of nomadic lifestyle to exploit different resources), as the challenging climatic conditions would have made food production unfeasible; thus, we would expect these population to remained at constant sizes through time. To our knowledge, the importance of the availability of local resources to the adoption of food production has never been tested with genetic data.
In this paper, we compare the demographic trajectories of populations with different lifestyles based on two datasets (Fig. 1).
Dataset 1: We gathered from the literature genome-wide data (>150 k SNPs) for 64 populations22,23,24,25,26,27,28,29,30,31,32,33,34 and estimated their demographic trajectories through time using the approach based on Linkage Disequilibrium (LD) developed by Mc Evoy and colleagues35 using the software NeON46. The populations were selected to avoid highly urbanized samples, and cover three major regions (sub-Saharan Africa, 28 populations, later referred to as Africa), southeastern Asia and Oceania (17 populations, later referred to as SE Asia), and Russia and Siberia (later referred to as Siberia, 19 populations).
Dataset 2: Pre-computed MSMC36 demographic estimates have been made available in a recent paper by Pagani and colleagues37. From this second set of data we selected 24 populations from the three already mentioned regions: Africa (5 populations), SE Asia (9 populations) and Siberia (10 populations).
We ask whether populations who turned to food production differ from hunter-gathers in their demography, and date these differences based on the estimates from the two different methods. We then proceed to test the extent to which these differences might be a direct consequence of resource availability from the surrounding environment using global paleoclimate and palaeovegetation reconstructions. Finally, to evaluate whether some demographic phenomena may generate patterns similar to those that we observed, in SE Asia we compare multiple migration scenarios that might have had a confounding effect over our inferences.
For each region, we computed the ratio of population sizes among all pairs with a different subsistence regime (NeFP/NeHG, Fig. 2a) to investigate formally how hunter-gatherers and food producers differed through time: a ratio of 1 would indicate no significant differences between the two categories. For all three regions and both datasets, consistently food producers had larger populations than hunter-gatherers (minimum ratio = 1.1). This difference was already detectable 20k years ago (17 k years ago in SE Asia, dataset 2), i.e. long before the inception of food-producing activities, and increased through time. The detailed trajectories of each population are presented in the Supplementary Figures 1 and 2.
We then tested whether these differences might be linked to climate. This explanation seems unlikely for Africa and SE Asia, as estimates of annual Net Primary Productivity (NPP) for these populations (assuming that they inhabited the same regions as in present times) were not consistently skewed in favor of future food producers (Fig. 2b). In Siberia, on the other hands, food producers inhabit areas that became progressively more favorable compared to those where hunter-gatherers are found. Indeed, if we quantify the number of effective individuals per unit of productivity (Ne/NPP), we see that the ratio for food producers versus hunter-gatherers was above 1 well before the advent of food production (Fig. 2c). For Africa and SE Asia, this ratio increased markedly around 10–12 k years ago, implying either an increase in the number of effective individuals sustained by the same amount of resources, or some immigration, while for Siberia the ratio remained flat through time.
Test for bottleneck in hunter-gatherers
The comparison between NeON trajectories calculated for American and European populations suggest that this method could underestimate Ne prior to a bottleneck46. Most modern-day hunter-gatherers are likely to have undergone repeated phenomena of fragmentation and/or demographic crisis, and, if so, our results could reflect to an extent that we cannot quantify a methodological bias. In other words, as previously observed6, based on measures of genetic diversity one may not be able to discriminate between long-term small population sizes, and recent bottlenecks affecting an originally large population.
MSMC is more robust to bottlenecks, and a comparison between the trajectories estimates with NeON and MSMC for the 12 populations that are shared between the two datasets shows that individual trajectories do indeed appear different when investigated by different methods, but this happens for all lifestyles, not only foragers (Supplementary Dataset 1), and the overall pattern when comparing lifestyles remains the same in both datasets.
An important assumption of our approach is that the populations in our study lived at approximately the same location over the last 20 k years, thus discounting the possibility of long-distance migrations. We tested the effect of this assumption for SE Asia, where the population analyzed have been suggested to derive from at least two waves of advance29,37,38,39. The more recent dispersal, the so-called Austronesian expansion, is documented in the archaeological record. It is interpreted as a spread of food producers from continental eastern Asia associated with the diffusion of Neolithic cultures and technologies, starting between 6,000 and 4,000 years ago2,40,41. Two main routes have been proposed for it: under the “Fast train” model42 the expansion started from China and spread through Taiwan reaching then island South-East Asia and Oceania. The “Slow boat” scenario43, instead, postulates a spread from Mainland South-East Asia.
To take into account the mentioned hypotheses, we calculated the ratio of Ne in SE Asia Dataset 1 following three models: “Cultural diffusion” (absence of migration, as presented in the main results section), “Fast Train” and “Slow Boat”. Under the cultural diffusion model, we associated to each Austronesian population the NPP of the region where they are now living (as we had done in the previous analyses). Under the “Fast Train” and “Slow Boat” models, we associated to them the NPP of the source region, respectively China and Cambodia.
As shown in Fig. 3, the overall pattern indicating an increase in more recent times does not vary much between models, showing that long distance migration does not seem to affect strongly our results. The main difference observed between models is that the minimum ratio is higher following the “Fast train” scenario. The reason is that the estimates of NPP for Eastern Asia are much lower than the ones observed in Taiwan and south-eastern Asia: as a consequence, the number of effective individuals per unit of NPP becomes much higher for food producers if based in a more temperate region such as China compared to more tropical areas.
Our global panel of populations revealed marked differences in inferred Ne between food producers and hunter-gatherers: as expected, the latter show larger effective population size. What was less obviously expected is that those differences began to accumulate 20 k years ago in all three regions, becoming more marked ~12–10 k years ago in Africa and SE Asia. It has to be stressed that these dates have been obtained using a generation time of 25 years44, which is considered an underestimate by some authors45, but allows direct comparisons with other studies. Had we chosen 30 years (e.g. as in ref.37), our time estimates for demographic growth would move even further back in time.
It is important to highlight that demographic estimates may suffer from different kinds of biases6. The method developed by McEvoy and colleagues appears to underestimate the Ne of non-Africans prior to the Out of Africa35,39, showing either a problem when dealing with bottlenecks (as suggested also by estimates on American populations46), or the confounding effect of population subdivision47. Even MSMC is not immune from the latter: the coalescent rate shows the same type of change when the population shrinks, and when a population of constant size gets subdivided48.
Neither possibility can be excluded, namely that future food producers already had larger population sizes before the inception of agriculture, or that hunter-gatherers have undergone repeated phenomena of fragmentation and/or demographic crisis. We tend to support the former view because, comparing estimates from different methods, it does not appear that only foragers have been subjected to such phenomena, possibly because most of the populations analysed are currently small and isolated groups of anthropological interest, rather than urban groups.
A certain level of uncertainty, at present, seems impossible to eliminate, and this is the reason why we prefer not to discuss the specific demographic reconstructions, but to focus instead on the comparison between lifestyles. Indeed, even if the individual trajectories may vary between NeON and MSMC, both datasets show the same signal of a difference between foragers and food producers that started before food production in agricultural communities. Moreover, our results match what has been already suggested based on other regions, markers and methods11,13,14,15,16,17,18,19, which makes us confident that the signal in the data is not dependent on the particular set of SNPs, or populations, or statistics chosen.
Whilst there are consistent differences between lifestyles across the three major regions covered by our study, Siberia stands out for showing a clear effect of climate. In this region, the increase in the difference between the two lifestyles can be mostly ascribed to food producers living in areas where the environment ameliorated markedly after the Last Glacial Maximum compared to more challenging conditions encountered by those populations that remained hunter-gatherers.
In other regions, on the other hand, we could not detect any effect of climate. We should emphasize that our reconstructions would only recover the medium-scale climatic conditions (in the order of 100 s of kilometers) that were encountered by a population, and would not be able to capture the effects of differences in the availability of local resources with a patchy distribution in space (such as freshwater from local rivers). Such local resources might well have played an important role for a number of populations; access to localized high-value resources has also been argued to develop societal structures that favor ownership and territory defense, and could have predisposed certain groups to be more likely to take up food production. On the other hand, the rough geographic resolution makes this method robust to short-distance migration, while the effects of long-distance migrations have been explicitly taken into account with the test performed on SE Asian populations, and is unlikely to account for the early dates of demographic growth inferred from the data.
The individual demographic trajectories in many cases show a decrease in the chronological window between 10 and 4 kya, corresponding to the establishment of food production in the analysed regions (Supplementary Figures 1 and 2), in contrast with archaeological data suggesting a demographic expansion following the Neolithic transition. Whilst the reasons for these declines remain unclear, they are in line with previous analyses, whether based on patterns of linkage disequilibrium on SNPchip data (e.g. ref.35), or on whole genomes analysed by PSMC (e.g. ref.37). The patterns found in our analysis seem then robust, as they are consistent with analyses based on different datasets and methods.
Apparent declines in Ne do not mean that the overall population was necessarily shrinking in size; indeed, the effective population size is affected by a variety of factors, including sex ratio, marriage patterns between and within groups, immigration, etc. Whatever the reason of this apparent decline could be, our analyses show that effective population sizes began to become larger in the ancestors of today’s food producers than in the ancestors of today’s foragers before the Neolithic transition (Fig. 2a). It seems more than likely that without the development of a new and, in the long run, more efficient subsistence technology, such an increase could not have lasted; agriculture doubtless created the resources to sustain larger populations. However, the demographic changes identified in our study cannot be regarded as a mere consequence of the increased food availability, but rather as a process preceding, and possibly stimulating, the Neolithic technological developments.
An early increase in Ne that predates Neolithization has been interpreted as capturing early societal changes that might have favored the later development of food production49. High population density can facilitate technical innovation, and populations in more advantageous areas that sustained higher densities might have led to the later improvement in subsistence technologies50. Moreover, in modern-day hunter-gatherers a larger population relative to ecological productivity is positively correlated to complex behaviors such as sedentism, storage activity and social stratification51.
Furthermore, indirect estimates of Ne from genetic data can also reflect immigration to an extent that can hardly be predicted, (with migration among previously isolated populations increasing Ne); areas where movement among populations and more connected networks of potential innovators might favor the development of food production in a manner similar to larger overall populations. Thus, large estimated Ne values might not represent just a large census size, but also high gene flow (and hence cultural connectivity), both of which could have favored innovation.
These two mechanisms are not mutually exclusive, and it is difficult to disentangle them genetically. However, the key result from our analysis is that, even when using a combination of genomes and a large amount of genome-wide data from a globally-distributed panel of populations, populations that later adopted food productions differed from those who remained hunter-gatherers well before their lifestyle changed. This process did not happen as result of differential resources but because of cultural, behavioral or social causes, maybe the same that have led to the major population replacement in Europe when hunter-gatherers and farmers met27,52,53,54. The very limited number of modern-day foragers from Western Eurasia, and the lack of genetic data from them, do not allow a direct test with the approach presented here, but similar results on European populations have been obtained with other methods19. Therefore, we conclude that pre-existing cultural or demographic differences among Paleolithic hunter-gatherers in the Old World likely played a role in the later choice of adopting food production.
Materials and Methods
Dataset 1: We compiled an extensive dataset of publicly available SNP data. We analyzed populations of hunter-gatherers and food producers from three regions in which both lifestyles are present in modern times (Fig. 1): sub-Saharan Africa (28 populations), southeastern Asia and Oceania (17 populations), Siberia (19 populations) (more information can be found in Supplementary Table 1). Only populations with a minimum of 15 individuals (10 for south-eastern Asia and Oceania) were considered, giving a total of >1200 individuals. The minimum number of SNPs used for the analyses for any given population was 150 k.
Dataset 2: MSMC demographic estimates for a large panel of worldwide populations have been published in the supplementary material of a recent publication by Pagani and colleagues37. The choice of populations is more limited, but populations with different lifestyles were available for all the three regions where SNP data were collected: Africa (5 populations) SE Asia (9 populations) and Siberia (10 populations) (Supplementary Table 2).
Lifestyle information for each population, when not available in the original reference, was recovered from Levinson (1991)55. Many Siberian populations adopt a variety of subsistence strategies, and in those cases, we classified them based on their primary activity. Geographic locations, when not available in the original references, where calculated as the center or the capital of the country where the sampling has been performed.
Estimation of Ne through time for Dataset 1
For each population, we used the pattern of Linkage Disequilibrium (LD) to estimate changes in Ne through time using the approach by McEvoy et al.35 as implemented in the R package NeON46. Given a known recombination rate, the amount of linkage disequilibrium (LD) between differently spaced loci can help reconstruct past values of Ne56. The reason is that smaller Ne leads to higher genetic drift, and hence to increased LD values. However, the greater the recombination rate between pairs of genetic markers, the faster the decay of LD between them. Since recombination accumulates through time, LD over large recombination distances gives an estimate of Ne in recent times, while LD over short recombination distances is informative on ancient Ne57.
NeON calculates the recombination rates for each possible pair of markers taking their genetic distance into account. We retrieved genetic maps of the human genome from the HapMap website (https://www.ncbi.nlm.nih.gov/probe/docs/projhapmap/), and the SNPs available for each population were mapped accordingly. Markers which could not be located on the HapMap maps were discarded.
In NeON, estimates of Ne are obtained by first assigning pairs of markers into several classes as a function of the recombination distance between them, and then calculating the squared correlation coefficient of linkage disequilibrium (r2LD)58. The r2LD is then used to estimate the value of effective population size within each of the identified categories, which, as discussed above, corresponds to the effective population size at a specific moment in the past.
Comparing hunter-gatherers and foragers
For both datasets we summarized the demographic estimates by computing the harmonic mean of Ne every 1,000 years from 20,000 until 4,000 years ago, using a generation time of 25 years44 (while in the original publication for Dataset 2 they use a generation time of 30 years)37. Given the low number of foragers in dataset 2, we considered as foragers not only hunter-gatherers but also horticulturalists, that in the analyses of dataset 1 are considered food producers.
For each region, we calculated the ratio between values of Ne in each possible pair of populations with different lifestyles (food producers over foragers, NeFP/NeHG). We then plotted the median of the ratio and calculated the error as the 95% distribution of the jackknife leave-one-out validation. A ratio of 1 would mean that the two different lifestyles have, on average, the same Ne.
Differences in Ne could be linked to a variety of environmental factors, such as climate and environmental productivity. To quantify this effect, we extracted Net Primary Productivity (NPP) estimates from paleoclimatic reconstructions59. We explored the changes in resource availability between populations with difference lifestyles by plotting with the same method described above for Ne the ratio between the estimates of NPP for the two lifestyles considered (NPPFP/NPPHG).
Finally, to correct for environmental effects on effective population size, we normalised Ne by NPP (number of effective individuals per unit of primary productivity). We again used the same method to calculate the median and 95% CI of Ne/NPP for food producers over hunter-gatherers. A ratio of 1 would mean that, once corrected for NPP, populations with different subsistence use the natural resources with the same efficiency.
Test of the method: Integrating long-distance migration in SE Asia
The approach employed in this paper ignores the effect of long distance migrations. This assumption is clearly unrealistic. To test how much the results could change if such migrations were taken into account, we modelled the Austronesian expansion in south-eastern Asia and Oceania under both the “Fast train”42 and the “Slow Boat”43 models. We then compared the results with the ones issued under the “Cultural Diffusion” model (absence of population movements).
For each model, Ne/NPP was calculated in a different way. Under the cultural diffusion model, we associated to each Austronesian population the NPP of the region where they are now living (the same logic used in the earlier analyses). Under the “Fast Train” and “Slow Boat” models, we associated to them NPP of the source region, respectively China and Cambodia.
Also in sub-Saharan Africa, there is evidence of a massive migration, the so-called “Bantu expansion”. However, since only one out of 28 African populations in our dataset belongs to the Bantu, this migration would have had a negligible effect on the analyses presented in this paper.
Armelagos, G. J., Goodman, A. H. & Jacobs, K. H. The origins of agriculture: Population growth during a period of declining health. Popul. Environ. 13, 9–22 (1991).
Bellwood, P. et al. First Farmers: the Origins of Agricultural Societies. Cambridge Archaeol. J. 17, 87 (2007).
Kılınç, G. M. et al. The Demographic Development of the First Farmers in Anatolia. Curr. Biol. 0, 137–140 (2016).
Bocquet-Appel, J.-P. When the world’s population took off: the springboard of the Neolithic Demographic Transition. Science 333, 560–561 (2011).
Harpending, H. & Rogers, A. Genetic perspectives on human origins and differentiation. Annu. Rev. Genomics Hum. Genet. 1, 361–385 (2000).
Excoffier, L. & Schneider, S. Why hunter-gatherer populations do not show signs of pleistocene demographic expansions. Proc. Natl. Acad. Sci. USA 96, 10597–602 (1999).
Destro-Bisol, G. et al. Variation of female and male lineages in sub-Saharan populations: the importance of sociocultural factors. Mol. Biol. Evol. 21, 1673–82 (2004).
Pilkington, M. M. et al. Contrasting signatures of population growth for mitochondrial DNA and Y chromosomes among human populations in Africa. Mol. Biol. Evol. 25, 517–25 (2008).
Aimé, C. et al. Microsatellite data show recent demographic expansions in sedentary but not in nomadic human populations in Africa and Eurasia. Eur. J. Hum. Genet. 22, 1201–1207 (2014).
Patin, E. et al. The impact of agricultural emergence on the genetic history of African rainforest hunter-gatherers and agriculturalists. Nat. Commun. 5 (2014).
Atkinson, Q. D., Gray, R. D. & Drummond, A. J. mtDNA variation predicts population size in humans and reveals a major Southern Asian chapter in human prehistory. Mol. Biol. Evol. 25, 468–74 (2008).
Batini, C. et al. Insights into the demographic history of African Pygmies from complete mitochondrial genomes. Mol. Biol. Evol. 28, 1099–110 (2011).
Zheng, H.-X. et al. Major population expansion of East Asians began before neolithic time: evidence of mtDNA genomes. PLoS One 6, e25835 (2011).
Zheng, H.-X., Yan, S., Qin, Z.-D. & Jin, L. MtDNA analysis of global populations support that major population expansions began before Neolithic Time. Sci. Rep. 2, 745 (2012).
Aimé, C. & Austerlitz, F. Different kinds of genetic markers permit inference of Paleolithic and Neolithic expansions in humans. Eur. J. Hum. Genet. 25, 360–365 (2017).
Chaix, R., Austerlitz, F., Hegay, T., Quintana-Murci, L. & Heyer, E. Genetic traces of east-to-west human expansion waves in Eurasia. Am. J. Phys. Anthropol. 136, 309–17 (2008).
Batini, C. et al. Large-scale recent expansion of European patrilineages shown by population resequencing. Nat. Commun. 6, 7152 (2015).
Aime, C. et al. Human Genetic Data Reveal Contrasting Demographic Patterns between Sedentary and Nomadic Populations That Predate the Emergence of Farming. Mol. Biol. Evol. 30, 2629–2644 (2013).
Maisano Delser, P. et al. Signatures of human European Palaeolithic expansion shown by resequencing of non-recombining X-chromosome segments. Eur. J. Hum. Genet. https://doi.org/10.1038/ejhg.2016.207 (2017).
Bar-Yosef, O. The Natufian culture in the Levant, threshold to the origins of agriculture. Evol. Anthropol. Issues, News, Rev 6, 159–177 (1998).
Berger, J. F. & Guilaine, J. The 8200 cal BP abrupt environmental change and the Neolithic transition: A Mediterranean perspective. Quat. Int. 200, 33–49 (2009).
Pagani, L. et al. Ethiopian genetic diversity reveals linguistic stratification and complex influences on the Ethiopian gene pool. Am. J. Hum. Genet. 91, 83–96 (2012).
Bryc, K. et al. Genome-wide patterns of population structure and admixture in West Africans and African Americans. Proc. Natl. Acad. Sci. USA 107, 786–91 (2010).
Li, J. Z. et al. Worldwide human relationships inferred from genome-wide patterns of variation. Science 319, 1100–4 (2008).
Schlebusch, C. M. et al. Genomic variation in seven Khoe-San groups reveals adaptation and complex African history. Science 338, 374–9 (2012).
Henn, B. M. et al. Hunter-gatherer genomic diversity suggests a southern African origin for modern humans. Proc. Natl. Acad. Sci. USA 108, 5154–62 (2011).
Lazaridis, I. et al. Ancient human genomes suggest three ancestral populations for present-day Europeans. Nature 513, 409–413 (2014).
Reich, D. et al. Denisova admixture and the first modern human dispersals into Southeast Asia and Oceania. Am. J. Hum. Genet. 89, 516–528 (2011).
Rasmussen, M. et al. An Aboriginal Australian genome reveals separate human dispersals into Asia. Science 334, 94–8 (2011).
Chaubey, G. et al. Population genetic structure in Indian Austroasiatic speakers: the role of landscape barriers and sex-specific admixture. Mol. Biol. Evol. 28, 1013–24 (2011).
Raghavan, M. et al. Upper Palaeolithic Siberian genome reveals dual ancestry of Native Americans. Nature 505, 87–91 (2014).
Yunusbayev, B. et al. The genetic legacy of the expansion of Turkic-speaking nomads across Eurasia. PLoS Genet. 11, e1005068 (2015).
Rasmussen, M. et al. Ancient human genome sequence of an extinct Palaeo-Eskimo. Nature 463, 757–62 (2010).
Fedorova, S. A. et al. Autosomal and uniparental portraits of the native populations of Sakha (Yakutia): implications for the peopling of Northeast Eurasia. BMC Evol. Biol. 13, 127 (2013).
McEvoy, B. P., Powell, J. E., Goddard, M. E. & Visscher, P. M. Human population dispersal ‘Out of Africa’ estimated from linkage disequilibrium and allele frequencies of SNPs. Genome Res. 21, 821–9 (2011).
Schiffels, S. & Durbin, R. Inferring human population size and separation history from multiple genome sequences. Nat. Genet. 46, 919–25 (2014).
Pagani, L. et al. Genomic analyses inform on migration events during the peopling of Eurasia. Nature 538, 238–242 (2016).
Lahr, M. M. & Foley, R. Multiple dispersals and modern human origins. Evol. Anthropol. Issues, News, Rev 3, 48–60 (1994).
Tassi, F. et al. Early modern human dispersal from Africa: genomic evidence for multiple waves of migration. Investig. Genet. 6, 13 (2015).
Reich, D., Thangaraj, K., Patterson, N., Price, A. L. & Singh, L. Reconstructing Indian population history. Nature 461, 489–94 (2009).
Pugach, I., Delfin, F., Gunnarsdóttir, E., Kayser, M. & Stoneking, M. Genome-wide data substantiate Holocene gene flow from India to Australia. Proc. Natl. Acad. Sci. USA 110, 1803–8 (2013).
Bellwood, P. S. Man’s conquest of the Pacific: the prehistory of Southeast Asia and Oceania. (Oxford University Press, 1979).
Solheim, W. G. The Nusantao and north-south dispersals. Bull. Indo-Pacific Prehistory Assoc. 15, 101–109 (1996).
Fenner, J. N. Cross-cultural estimation of the human generation interval for use in genetics-based population divergence studies. Am. J. Phys. Anthropol. 128, 415–23 (2005).
Tremblay, M. & Vézina, H. New Estimates of Intergenerational Time Intervals for the Calculation of Age and Origins of Mutations. Am. J. Hum. Genet. 66, 651–658 (2000).
Mezzavilla, M. & Ghirotto, S. Neon: An R Package to Estimate Human Effective Population Size and Divergence Time from Patterns of Linkage Disequilibrium between SNPS. J. Comput. Sci. Syst. Biol. 8, 037–044 (2015).
Eriksson, A. & Manica, A. The doubly conditioned frequency spectrum does not distinguish between ancient population structure and hybridization. Mol. Biol. Evol. 31, 1618–21 (2014).
Mazet, O., Rodríguez, W., Grusea, S., Boitard, S. & Chikhi, L. On the importance of being structured: instantaneous coalescence rates and human evolution—lessons for ancestral population size inference? Heredity (Edinb). 116, 362–371 (2016).
Sauer, C. O. Agricultural origins and dispersals. (American Geographical Society, 1952).
Powell, A., Shennan, S. & Thomas, M. G. Late Pleistocene demography and the appearance of modern human behavior. Science 324, 1298–301 (2009).
Rowley-Conwy, P. In Hunter-gatherers: an interdisciplinary perspective (eds Panter-Brick, C., Layton, R. H. & Rowley-Conwy, P.) 39–72 (Cambridge University Press, 2001).
Fu, Q. et al. The genetic history of Ice Age Europe. Nature, https://doi.org/10.1038/nature17993 (2016).
Hofmanová, Z. et al. Early farmers from across Europe directly descended from Neolithic Aegeans. Proc. Natl. Acad. Sci. USA 113, 6886–91 (2016).
Skoglund, P. et al. Genomic diversity and admixture differs for Stone-Age Scandinavian foragers and farmers. Science 344, 747–50 (2014).
Levinson, D. Encyclopedia of world cultures. (Boston, Mass, G.K. Hall, 1991).
Hill, W. G. & Robertson, A. Linkage disequilibrium in finite populations. Theor. Appl. Genet. 38, 226–31 (1968).
Hayes, B. J., Visscher, P. M., McPartlan, H. C. & Goddard, M. E. Novel multilocus measure of linkage disequilibrium to estimate past effective population size. Genome Res 13, 635–43 (2003).
Awadalla, P., Eyre-Walker, A. & Smith, J. M. Linkage disequilibrium and recombination in hominid mitochondrial DNA. Science 286, 2524–5 (1999).
Eriksson, A. et al. Late Pleistocene climate change and the global expansion of anatomically modern humans. Proc. Natl. Acad. Sci. USA 109, 16089–94 (2012).
QGIS Development Team. QGIS Geographic Information System. Open Source Geospatial Foundation Project. http://www.qgis.org/ (2017).
We would like to thank Silvia Ghirotto and Francesca Tassi for their help and useful comments. We are very grateful to the two anonymous reviewers whose comments and suggestions greatly improved the quality of the paper. This work was supported by the European Research Council ERC-2011-AdG_295733 grant (LanGeLin) to GB and ERC Consolidator Grant 647787 ‘LocalAdaptation’ to AM. ML has been awarded with the SIBE (Società Italiana di Biologia Evolutiva) “Doctor Darwin prize” for this research.
The authors declare that they have no competing interests.
Change History: A correction to this article has been published and is linked from the HTML version of this paper. The error has not been fixed in the paper.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
A correction to this article is available online at https://doi.org/10.1038/s41598-017-18588-0.
Electronic supplementary material
About this article
Cite this article
Leonardi, M., Barbujani, G. & Manica, A. An earlier revolution: genetic and genomic analyses reveal pre-existing cultural differences leading to Neolithization. Sci Rep 7, 3525 (2017). https://doi.org/10.1038/s41598-017-03717-6