A western route of prehistoric human migration from Africa into the Iberian Peninsula.

Being at the western fringe of Europe, Iberia had a peculiar prehistory and a complex pattern of Neolithization. A few studies, all based on modern populations, reported the presence of DNA of likely African origin in this region, generally concluding it was the result of recent gene flow, probably during the Islamic period. Here, we provide evidence of much older gene flow from Africa to Iberia by sequencing whole genomes from four human remains from northern Portugal and southern Spain dated around 4000 years BP (from the Middle Neolithic to the Bronze Age). We found one of them to carry an unequivocal sub-Saharan mitogenome of most probably West or West-Central African origin, to our knowledge never reported before in prehistoric remains outside Africa. Our analyses of ancient nuclear genomes show small but significant levels of sub-Saharan African affinity in several ancient Iberian samples, which indicates that what we detected was not an occasional individual phenomenon, but an admixture event recognizable at the population level. We interpret this result as evidence of an early migration process from Africa into the Iberian Peninsula through a western route, possibly across the Strait of Gibraltar.

Demographic history and adaptation account for clock gene diversity in humans.

Circadian clocks give rise to daily oscillations in behavior and physiological functions that often anticipate upcoming environmental changes generated by the Earth rotation. In model organisms a relationship exists between several genes affecting the circadian rhythms and latitude. We investigated the allele distributions at 116 000 single-nucleotide polymorphisms (SNPs) of 25 human clock and clock-related genes from the 1000Genomes Project, and at a reference data set of putatively neutral polymorphisms. The global genetic structure at the clock genes did not differ from that observed at the reference data set. We then tested for evidence of local adaptation searching for FST outliers under both an island and a hierarchical model, and for significant association between allele frequencies and environmental variables by a Bayesian approach. A total of 230 SNPs in 23 genes, or 84 SNPs in 19 genes, depending on the significance thresholds chosen, showed signs of local adaptation, whereas a maximum of 190 SNPs in 23 genes had significant covariance with one or more environmental variables. Only two SNPs from two genes (NPAS2 and AANAT) exhibit both elevated population differentiation and covariance with at least one environmental variable. We then checked whether the SNPs emerging from these analyses fall within a set of candidate SNPs associated with different chronotypes or sleep disorders. Correlation of five such SNPs with environmental variables supports a selective role of latitude or photoperiod, but certainly not a major one.

Nine things to remember about human genome diversity.

Understanding how and why humans are biologically different is indispensable to get oriented in the ever-growing body of genomic data. Here we discuss the evidence based on which we can confidently state that humans are the least genetically variable primate, both when individuals and when populations are compared, and that each individual genome can be regarded as a mosaic of fragments of different origins. Each population is somewhat different from any other population, and there are geographical patterns in that variation. These patterns clearly indicate an African origin for our species, and keep a record of the main demographic changes accompanying the peopling of the whole planet. However, only a minimal fraction of alleles, and a small fraction of combinations of alleles along the chromosome, is restricted to a single geographical region (and even less so to a single population), and diversity between members of the same population is very large. The small genomic differences between populations and the extensive allele sharing across continents explain why historical attempts to identify, once and for good, major biological groups in humans have always failed. Nevertheless, racial categorization is all but gone, especially in clinical studies. We argue that racial labels may not only obscure important differences between patients but also that they have become positively useless now that cheap and reliable methods for genotyping are making it possible to pursue the development of truly personalized medicine.

Comparing models on the genealogical relationships among Neandertal, Cro-Magnoid and modern Europeans by serial coalescent simulations.

Populations of anatomically archaic (Neandertal) and early modern (Cro-Magnoid) humans are jointly documented in the European fossil record, in the period between 40 000 and 25 000 years BP, but the large differences between their cultures, morphologies and DNAs suggest that the two groups were not close relatives. However, it is still unclear whether any genealogical continuity between them can be ruled out. Here, we simulated a broad range of demographic scenarios by means of a serial coalescence algorithm in which Neandertals, Cro-Magnoids and modern Europeans were either part of the same mitochondrial genealogy or of two separate genealogies. Mutation rates, population sizes, population structure and demographic growth rates varied across simulations. All models in which anatomically modern (that is, Cro-Magnoid and current) Europeans belong to a distinct genealogy performed better than any model in which the three groups were assigned to the same mitochondrial genealogy. The maximum admissible level of gene flow between Neandertals and the ancestors of current Europeans is 0.001% per generation, one order of magnitude lower than estimated in previous studies not considering genetic data on Cro-Magnoid people.

Autocorrelation of gene frequencies under isolation by distance.

Spatial autocorrelation statistics are used for description of geographic variation of gene frequencies, but the relationship of these indices with the parameters describing the genetic structure of populations is not established. A simple relation is derived here between kinship coefficient and a measure of spatial autocorrelation, Moran's I. The autocorrelation coefficient of gene frequencies at a given distance is a direct function of the kinship at that distance, and an inverse function of the standardized gene frequency variance, Fst. Under isolation by distance, the expected values of Moran's I for any allele may be calculated by means of Malécot-Morton function, which predicts an exponential decline of genetic similarity in space. This allows comparison of observed gene frequency patterns with the patterns that should be caused by interaction of short range migration and random genetic drift.

Analysis of DNA diversity by spatial autocorrelation.

Two statistics are proposed for summarizing spatial patterns of DNA diversity. These autocorrelation indices for DNA analysis, or AIDAs, can be applied to RFLP and sequence data; the resulting set of autocorrelation coefficients, or correlogram, measures whether, and to what extent, individual DNA sequences or haplotypes resemble the haplotypes sampled at arbitrarily chosen spatial distances. Analyses of computer-generated sets of data, and of RFLP data from two natural populations, show that AIDAs allow one to objectively and simply identify basic patterns in the spatial distribution of haplotypes. These statistics, therefore, seem to be a useful tool both to explore the genetic structure of a population and to suggest hypotheses on the evolutionary processes that shaped the observed patterns.

Molecular polymorphism in the period gene of Drosophila simulans.

The threonine-glycine (Thr-Gly) repeat region of the period (per) gene of eight natural populations of Drosophila simulans from Europe and North Africa was analyzed by polymerase chain reaction, DNA sequencing and heteroduplex formation. Five different length alleles encoding 21, 23, 25 and two different kinds of 24 Thr-Gly pairs in the uninterrupted repeat were found. In the 3' region flanking the repeat 6 nucleotide substitutions (3 synonymous, 3 replacement) were observed in three different combinations that we called haplotypes I, II and III. The complete linkage disequilibrium observed between the haplotypes and these length variants allowed us to infer from the repeat length, the DNA sequence at the 3' polymorphic sites. The haplotypes were homogeneously distributed across Europe and North Africa. The data show statistically significant departures from neutral expectations according to the Tajima test. The results suggest that balancing selection might have played a role in determining the observed levels and patterns of genetic diversity at the per gene in D. simulans.

Do Basque- and Caucasian-speaking populations share non-Indo-European ancestors?

Genetic evidence is consistent with the view that the Indo-European languages were propagated in Europe by the diffusion of early farmers. The existence of phylogenetic relationships between European populations speaking other languages has been proposed on linguistic and archaeological grounds, and is here tested by analyzing allele frequencies at ten polymorphic protein and blood group loci. Genetic distances between speakers of Basque and Caucasian languages are compared with those between controls, i.e. contiguous populations speaking Indo-European and Altaic. Although some statistical tests show an excess of genetic similarity between Basque and South Caucasian speakers, most results do not support their common origin. If the Basques and the Caucasian-speaking populations share common ancestors, recent evolutionary phenomena must have caused divergence between them, so that their gene frequencies do not appear more similar now than those of random pairs of populations separated by the same geographic distance.

High carrier frequency of the 35delG deafness mutation in European populations. Genetic Analysis Consortium of GJB2 35delG.

Congenital deafness accounts for about 1 in 1000 infants and approximately 80% of cases are inherited as an autosomal recessive trait. Recently, it has been demonstrated that connexin 26 (GJB2) gene is a major gene for congenital sensorineural deafness. A single mutation (named 35delG) was found in most recessive families and sporadic cases of congenital deafness, among Caucasoids, with relative frequencies ranging from 28% to 63%. We present here the analysis of the 35delG mutation in 3270 random controls from 17 European countries. We have detected a carrier frequency for 35delG of 1 in 35 in southern Europe and 1 in 79 in central and northern Europe. In addition, 35delG was detected in five out of 376 Jewish subjects of different origin, but was absent in other non-European populations. The study suggests either a single origin for 35delG somewhere in Europe or in the Middle East, and the possible presence of a carrier advantage together with a founder effect. The 35delG carrier frequency of 1 in 51 in the overall European population clearly indicates that this genetic alteration is a major mutation for autosomal recessive deafness in Caucasoids. This finding should facilitate diagnosis of congenital deafness and allow early treatment of the affected subjects.

Mitochondrial DNA sequences in prehistoric human remains from the Alps.

The spread of agriculture that started in the Near East about 10 000 years ago caused a dramatic change in the European archaeological record. It is still unclear if that change was caused mostly by movement of people or by cultural transformations. In particular, there is disagreement on what proportion of the current European gene pool is derived either from the pre-agricultural, paleolithic and mesolithic people, or from neolithic farmers immigrating from the south-east. To begin to characterise the mtDNA gene pool of prehistoric Europe we examined five human remains from the Eastern Italian Alps, dated between 14 000 and 3000 years ago. Three of them yielded sufficient amount of mtDNA for analysis. DNA extracts were prepared in two independent laboratories, and PCR products from the first hypervariable segment of the mtDNA control region were cloned and sequenced. Together with the 5200 year old 'ice man', these DNA sequences show that European mtDNA diversity was already high at the beginning of the neolithic period. All the neolithic sequences have been observed in contemporary Europeans, suggesting genealogical continuity between the neolithic and present-day European mtDNA gene pool. The mtDNA sequence from a 14 000 year-old specimen was not observed in any contemporary Europeans, raising the possibility of a lack of continuity between the mesolithic and present-day European gene pools.

Mitochondrial lineages in Ladin-speaking communities of the eastern Alps.

European mitochondrial alleles cluster into five haplogroups. Haplogroup 2 is rare in general, but represents more than half of the few known sequences among Ladin speakers of the Alps. Here we describe DNA diversity in control region I of the hypervariable D-loop in 43 Ladins, and in 25 Italian speakers. Analysis of these data, and of previously published sequences, confirms a high degree of differentiation among Ladins and their geographical neighbours. This cannot be regarded as a simple effect of isolating factors, geographic or linguistic, as diversity is high within Ladin communities too. Rather, allele genealogies, population trees, and principal component analysis suggest a relationship between Ladin and Near Eastern samples. Two evolutionary hypotheses seem compatible with these findings. The view whereby Ladins could be descended from Palaeolithic inhabitants of the Alps is supported by the identification, in this study, of the probable ancestral haplotype of group 2, never previously observed in central Europe. Alternatively, a comparatively recent, Neolithic immigration of the ancestors of current Ladin speakers seems consistent with recent linguistic theories. In both cases, the number of lineages present, and their extensive diversity, are not compatible with a serious bottleneck in the Ladin population's history.

A latitudinal cline in a Drosophila clock gene.

The clock gene period determines biological rhythmicity in Drosophila melanogaster and encodes a protein characterized by an alternating series of threonine-glycine pairs. The minisatellite region encoding the threonine-glycine repeat is polymorphic in length in natural Drosophila melanogaster populations. In this paper we report the geographical analysis of this polymorphism within Europe and North Africa. A robust clinal pattern is observed along a north-south axis. We suggest the possibility that the length polymorphism could be maintained by thermal selection because the threonine-glycine region has been shown to provide thermostability to the circadian phenotype.

Linkage of biopsy, cancer, and population records aimed at the estimation of family risks in neoplasia: a pilot study.

STUDY OBJECTIVE: The aim was to link individual demographic and medical records into sibships to obtain the sibling distribution of biopsies and cancers, and thereby calculate heritability and recurrence risks in families, thus aiding early diagnosis and prevention of cancers. DESIGN: The 157,823 individual records of the inhabitants of the town of Ferrara in Italy were automatically linked into 106,821 sibships. A 10% sample (10,842 sibships) was then extracted from the distribution of sibships and tabulated, for linkage to medical records. PATIENTS: The biopsy records at the Institute of Pathological Anatomy of the University of Ferrara were manually linked to cancer records and then to sibships. It was possible to construct the distribution of 2062 biopsies and of 829 cancers in sibships. RESULTS: From the distribution of biopsies and tumours in sibships, it was possible to estimate the incidence of tumours in the population (0.052) and in siblings of affected (0.083), and to apply to such distributions current methods for the estimate of heritability (h2 = 0.246) and of recurrence risks of tumours in sibships, age independent. CONCLUSIONS: The study shows that the procedure resulting in the estimation of incidences and recurrence risks for tumours could be completely automated, and extended to whole populations and homogeneous subgroups in post industrial cultures.

What do languages tell us about human microevolution?

Analysis of cultural traits, especially from linguistic data, is increasingly being used to interpret gene-frequency variation among human populations. Evolutionary processes are inferred using two major approaches, one based on reconstruction of demographic events in time, and the other based on location of factors of evolutionary relevance in space. This review discusses some assumptions underlying these approaches, and suggests that they may lead to different, although not incompatible, conclusions.

Diversity of some gene frequencies in European and Asian populations. IV. Genetic population structure assessed by the variogram.

Isolation-by-distance models of population structure predict an exponential decrease of genetic relatedness with distance. Under the Kimura-Weiss (1964) model the plots of gene frequency variance versus distance (variograms), computed at various loci, are expected to have equal slope and reach a common asymptote. The gene frequency distributions at eight loci in 192 European and Asian populations have been summarized by variograms. On the average, the Kimura-Weiss model seems to describe adequately allele frequency change up to 900 km, but gradients are apparent at greater distances for most markers studied. These patterns may result form either differential selection or long-range gene flow; however, the extensive clinal variation observed for glyoxalase, esterase D and 6-PGD cannot be entirely accounted for by the Neolithic radiation of early farmers in the Middle East and Europe.

A two-step test for the heterogeneity of Fst values at different loci.

Under genetic drift and in the absence of selection, the Fst values are expected to be equal at all loci, and heterogeneity among such values is considered as an evidence for different systematic pressures affecting the different genetic systems considered. A two-step test is here proposed. Fst values at various loci are classified into two nonarbitrary groups through the analysis of the association between pairs of genetic and environmental variables; the heterogeneity is then assessed by Wilcoxon's two-sample test. Estimation of the theoretical variance of Fst, which causes major shortcomings in the test proposed by Lewontin and Krakauer in 1973, is thus avoided.

A review of statistical methods for continuous monitoring of malformation frequencies.

Six statistical methods for monitoring of populations are reviewed with special reference to surveillance of congenital malformations. Their characteristics are compared; their relative efficiencies are assessed through both analytical and numerical comparisons. The cumulative sum technique appears to be convenient from most points of view; particular cases in which other methods may be preferred are discussed.

Diversity of some gene frequencies in European and Asian populations. III. Spatial correlogram analysis.

The gene frequencies at eight loci in some European and Asian human populations have been subjected to spatial autocorrelation analysis, using Geary's c coefficient. Contrary to what is expected for markers affected only by gene flow and genetic drift, the spatial correlograms show distinct modes of gene frequency variation: there are significant clinal patterns (at the GLO and ESD loci), significant non-clinal patterns (AK, ADA, 6-PGD and GPT) and marginally significant patterns (PGP and SOD). Any hypothesis on the evolution of these polymorphisms should account for the observed heterogeneity of their geographical distributions.

Geographic patterns: how to identify them and why.

Geographic patterns of genetic diversity allow us to make inferences about population histories and the evolution of inherited disease. The statistical methods describing genetic variation in space, such as estimation of genetic variances, mapping of allele frequencies, and principal components analysis, have opened up the possibility to reconstruct demographic processes whose effects have been tested by a variety of approaches, including spatial autocorrelation, cladistic analyses, and simulations. These studies have significantly contributed to our understanding of human genetic variation; however, the molecular data that have accumulated since the mid-1980s have also created new complications. Reasons include the generally limited sample sizes, but, more generally, it is the nature of molecular variation itself that makes it necessary to develop and apply specific models and methods for the treatment of DNA data. The foreseeable diffusion of laboratory techniques for the rapid typing of many DNA markers will force us to change our approach to the study of human variation anyway, moving from the gene level toward the genome level. Because extensive variation among loci is the rule rather than the exception, an important practical tip is to be skeptical of inferences based on single-locus diversity.