Y-DNA Haplogroup R1a

new

Admin
Administrator

Posts: 72,768

Y-DNA Haplogroup R1a Aug 27, 2014 21:05:57 GMT

Quote

Post by Admin on Aug 27, 2014 21:05:57 GMT

Eastern Europeans and British Asians from India and Pakistan share the same haplogroup called R1a-M17. The R1b tribe originally migrated to Europe from West Asia from 4,000 BCE and the Indo-Europeans have shared ancestry that can be traced back to North India, which is why the Spaniards and some Europeans without Scandinavian admixture are physically similar to Asians.

Genetic structure of the studied regional population groups
We observed a total of 19 Y-haplogroups in our analyzed dataset of 621 Y-chromosomes (Table 1) defined by 31 informative polymorphisms out of 55 genotyped binary polymorphisms (Supplementary Figure 1). It has been argued in the literature that the Indian higher caste groups show relatively small genetic distances when compared with the West Eurasians,11 linking this to hypothetical migrations by Indo-Aryan speakers. Further, M17-R1a (presently designated as R1a1) was suggested as a potential marker with decreasing frequencies from Central Asia towards South India.23 On similar lines, it was suggested that a package of Y-HGs (J2, R1a, R2 and L) was associated with the migration of Indo-European people from Central Asia.7 Although our study observed a high frequency of Y-HGs, R1a1, J*/J2, R2 and L, it was not exclusively restricted to any region or population (Table 1). Moreover, most of the population groups from the studied regions showed a less frequency of the highly frequent haplogroups of Central Asia: C3, DE, I, G, J*, N and O, except for some population-specific distributions. Y-haplogroup G was observed to be present at high frequency in Gujarat Brahmins and Bihar Paswans, whereas Y-haplogroup O was more frequent in Uttar Pradesh Kols and Gonds (Table 1). In case of a recent gene flow (associated with the migration of Indo-European people), we expected the more frequent Central Asian Y-haplogroups (C3, DE, I, G, J*, N and O)12 to be present at least at similar frequencies, as observed for R1a1* in Northwest India, which, however, was not the case in this study.

Comparison of Brahmins and scheduled castes/tribals
To explore further, we analyzed the dataset (consisting of 510 Y-chromosomes), which could be classified as Brahmins (n=256) and scheduled castes/tribes (n=254) from the studied six regions of India (Jammu and Kashmir, Uttar Pradesh, Bihar, Madhya Pradesh, Maharastra and Gujarat) to evaluate regional distribution patterns between these two extreme end population groups of the Hindu caste hierarchy (Supplementary Table 1), where intermixing due to marriages has been absent because of social unacceptability. AMOVA showed no variation between different geographical regions (−2.3%), some variation between populations within regions (12.67%) and most of the variation within populations (89.63%). The percentage distribution of haplogroups (Supplementary Table 1) in Brahmins (n=256) showed a total of six most frequent (percentage >5%) haplogroups: R1a1* (40.63%), J2 (12.5%), R2 (8.59%), L (7.81%), H1 (6.25%) and R1* (5.47%), contributing to 81.25% of the total distribution in Brahmins. Tribals and scheduled castes (n=254) also showed six haplogroups: H1 (31.10%), R1a1* (20.47%), J2 (10.24%), L (7.87%), H* (7.87%) and O (6.69%), contributing in total to 84.25%. Interestingly, four of the haplogroups were overlapping in percentage (>5%) distribution with Brahmins. The haplogroup diversity and s.d. in each population are also given in Supplementary Table 1.higher resolution of R1a1* and confirm the present conclusions.

Study of the compiled dataset
The pooled percentage distribution of Y-haplogroups in the overall dataset of 2809 Y-chromosomes (767 Brahmins, 674 schedule castes and 1368 tribals) is summarized in Supplementary Table 2. All together (Brahmins, schedule castes and tribals), 22 Y-haplogroups were observed. The percentages of seven of these haplogroups (with percentage >5%) accounted for 85.5% of the total number of Y-chromosomes (n=2809). The haplogroups with their percentages in descending order were: R1a1* (21.1%), H1 (19.1%), R2 (10.5%), O (10.1%), L (9.5%), J*/J2 (8.3%) and F* (6.9%). These haplogroups remained the most frequent haplogroups even after the distribution of Y-chromosomes within respective groups of Brahmins, schedule castes and tribals, but with significant percentage differences (Supplementary Table 2). Five haplogroups out of 18 were found to be most frequent (>5%) in Brahmins (R1a1* (35.7%), J*/J2 (12.4%), L (11.3%), R2 (10.8%) and H1 (8.0%)) and represented 78.2% of the total number of samples (n =767), whereas haplogroup O was found to be very less frequent (0.7%) in Brahmin Y-chromosomes. Seven out of 14 haplogroups (with percentage >5%) (H1 (24.2%), R1a1* (17.2%), R2 (14.2%), L (12.2%), F* (9.8%), J*/J2 (6.4%) and K* (5.3%)) represented 89.3% of the total number of Dalit Y-chromosomes (n =674). Tribal Y-chromosomes represented by seven out of 20 haplogroups displayed percentages >5%: O (25.5%), H1 (25.3%), R1a1* (10.2%), F* (7.5%), R2 (6.4%), J*/J2 (6.1%) and L (5%) (86% of the total number of samples (n=1368)). All other observed haplogroups had their percentages <5% (Supplementary Table 2). The study was further extended, dividing the samples into four main linguistic categories (Indo-European (IE), Dravidian (DR), Tibeto-Burman (TB) and Austro-Asiatic (AA)) present in India as well as five regional categories (Central, East, North, South and West India). Y-haplogroup distributions as per these categories are presented in Supplementary Figures 3a and b. AMOVA was also done using the compiled dataset and by characterizing the populations into social, geographical and linguistic groups (Table 2). Geographical regions showed very less variation (0.79%) among the groups but higher variation between populations within groups (16.94%). In contrast, linguistic groups showed higher variation among the groups (15.56%) but lower variation between populations within linguistic groups (6.15%). Interestingly, when the TB linguistic group was removed from the analysis, the percentage variation among the groups reduced (9.43%) but variation between populations remained almost the same (Table 2). It was observed that by either of the grouping most of the variation was within the population groups.

Figure 1. The spatial distribution maps of Y-haplogroup R1a1 generated by the Kriging procedure using SURFER version 8.0. (a) Spatial frequency distribution of Y-haplogroup R1a1* across Eurasia, Central Asia and the Indian subcontinent. (b) Spatial distribution of Y-haplogroup R1a1*-associated diversity based on microsatellite markers.

Origin of Y-haplogroup R1a1*
However, a peculiar trend in distribution of the highest frequency of Y-haplogroup R1a1* (Table 1) in Brahmins, H1 in tribals and schedule castes, and O in tribals was also observed. Whereas on the one hand a consensus has developed in the literature among all schools of thought in assigning Indian origin to haplogroup H1 and in the association of haplogroup O with either Austro-Asiatic or Tibeto-Burman tribals, the widespread geographic distribution of R1a1* and reasonably high frequency across Eurasia (Figure 1a), with scanty representation of its ancestral (R*, R1* and R1a*) and derived lineages (R1a1a, R1a1b and R1a1c) across the region, leaves obscure the question of origin of R1a1*. This becomes more complex with the claims7, 9, 12, 23 proposing a scenario of the recent major gene flow from Central Asia to India and the antagonistic observations9, 12 of its highest variance in India, suggesting the gene flow in opposite direction. Further, the observation of a very high frequency (upto 72.22%) in this study (Table 1) and in the literature (Supplementary Figures 3a and b) of this haplogroup in all of the Brahmins may indicate its presence as a founder lineage for this caste group (irrespective of the geographical and linguistic affiliation of Brahmins), thus making this haplogroup of extreme importance and a key haplogroup in answering the question of origin of caste systems in India.

Although the geographic origins of haplogroup expansions can be inferred from the frequencies, associated diversity51 and clinal patterns of distribution, past inferences from literature indicate that such relations are not so simple to interpret. It is observed that regions of high frequency and high variance are not always the same. Regions with highest haplogroup frequencies are not always sites of its origin and clinal patterns are not obvious in binary HG frequency data;33 also, the highly associated microsatellite variance, exclusively, may not always be an indicator of in-situ diversification and could result as a consequence of repeated gene flow from different sources52, 53 as observed by Y-chromosomal diversity in Central Asia.32 This suggests that many analytical parameters should be included and potential causes of a wrong interpretation should be taken care of before reaching any conclusion. All rival models of the origin of caste system were taken into consideration and results were analyzed to the highest Y-SNP marker resolution for the R1a haplogroup (Supplementary Figure 1), in addition to adding data and information from the literature, making a pooled dataset of the R1a1* haplogroup containing ~1030 individuals (530 Indians, 224 Pakistanis, and 276 Central Asians and Eurasians) from around the Indian-subcontinent, Central Asia and Europe. Using different phylogenetic tools and parameters (mentioned in Materials and methods) and concentrating on the distribution of R1a1* and its ancestral lineages, the answer to the source and expansion of this haplogroup, across the globe, was explored.

Spatial frequency and molecular diversity distribution of R1a1* in Eurasia, Central Asia and the Indian subcontinent
The spatial frequency distribution of R1a1* across Eurasia along with spatial representation of associated diversity based on microsatellite markers within the haplogroup are given in Figures 1a and b. It was interesting to find that by adding information regarding the frequency and diversity of R1a1* from different population groups of North India (Information from North Indian population groups was scanty in earlier publications from India.) to the pooled data from different published sources, a clearer picture emerged, with overlapping high frequency and molecular diversity of R1a1* within India.

Admixture and diversity analysis
Considering the very high frequency of R1a1* (upto 72.22% as in WB) in Brahmins, irrespective of their geographical and linguistic affiliations, admixture analysis41 based on pooled data was performed. Three models of potential parental contributions of R1a1* (Figure 2) were tested, to evaluate the concepts of Central Asian introduction of the Indian caste system7 by Indo-Aryans (appointing themselves to the castes of higher ranks),14 as well as of rank-related West Eurasian admixture.11, 21 The observed proportions of contributions, taking all populations (Europeans (EU), Central Asians (CA) and Indian Brahmins (IB)) alternatively as source populations under different models (Figure 2), suggested model 3 (CA+IB → EU) as the best fit model (tested by 1000 bootstraps) and model 2 also as a possibility, for contributions of R1a1*, based on both proportion of frequency distribution as well as molecular divergence. Admixture analysis in light of other genetic evidences from this study did not seem to favor either Central Asian origin of the haplogroup or rank-related Eurasian admixture; instead it supported the Indian origin of this haplogroup and its contributions to other regions.

Figure 2. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Admixture proportions were estimated using ADMIX2 software under different models. All populations (Europeans (EU), Central Asians (CA) and Indian Brahmins (IB)) were considered alternatively as source populations and the respective proportions of contributions were estimated. mY1 and mY2 are the estimated admixture coefficients, corresponding to the relative contribution to the hybrid population (Hyb) from the parental populations (P1 and P2, respectively).

Further, the average diversity of the R1a1* haplogroup in Central Asians, Europeans and Indians was also calculated. The highest diversity of 0.52 (for both sampling and stochastic processes s.d.=0.32) was observed in Indians when compared with Europeans (0.40, s.d.=0.27) and Central Asians (0.32, s.d.=0.23). The calculation of Spearman's rank correlation coefficients46 between the latitude and longitude with haplogroup R1a1* frequency (r2=−0.13, 0.30) did not show any significant correlation, The same observation for R1a1* diversity (r2=−0.25, 0.20) has been reported earlier as well.9 This observation is again in favor of the suggestion that there has been no bulk migration from Central Asia to India.

Molecular evidences for the origin of R1a1* in the Indian subcontinent
The median joining network40 was also constructed. This algorithm provides the best results when applied on datasets of multi-state markers but within closely related haplotypes54 as is the case, using pooled data of R1a1* haplogroup. The inferences from the analysis (Figure 3) were again in favor of our earlier observations. The Indian haplotypes were observed to be the most diverse, and haplotypes spanning Central Asia and Eurasia, along with some Indian regional haplotypes, seemed to be derived as a subset of this diversity. The extremely high level of sharing of haplotypes across the regions as well as reticulations, mostly with one step difference, in this subset suggests parallel evolution of different haplotypes, which appears more plausible after their geographical distribution and expansions. However, the diversity within the Indian populations, represented by the long branches and links connecting many haplotypes, is also an indicator of their ancestry, geographical differentiation and severe bottlenecks within India, suggesting loss of many of the intermediate haplotypes, thus reducing the reticulation and increasing the branches’ length. The observed genetic distances FST38 and 1−PSA44 within the R1a1* haplogroup, between Central Asians (CA), Europeans (EU), as well as pooled populations of the Indian subcontinent (IS) showed overlapping trends of distribution. FST is based on the total variance in allele frequencies among populations and 1−PSA considers shared allele frequencies. IS populations showed less sharing with the CA (FST=0.095, 1−PSA=0.61) as compared with the EU (FST=0.021, 1−PSA=0.73) populations. AMOVA for these three pooled population groups (EU, CA, IS) showed that 94.07% of the total variation is present within the population, whereas only 5.93% of the differences are observed among population groups.

Figure 3. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Median joining network based on Y-STR haplotypes within Y-haplogroup R1a1*, showing the relationship between Indian, Central Asian and Eurasian population groups. *Biallelic marker M17 was included with the highest weight. The root of the network represents an individual with SRY10831b-R1a* (x M17-R1a1).

Age estimates for Y-haplogroup R1a1*
The age of microsatellite variations was re-calculated using Y-STRs data and by applying mutation rates and generation times (discussed in Materials and methods) within R1a1* lineage in Central Asia, Eurasia, Pakistan, as well as Indian populations (Table 3), and compared with the already published ages. The ages of the haplogroup, within the various population groups of India as well as after distributing them to social groups, were also calculated (Table 3). It was observed that the age of R1a1* was the highest in the Indian subcontinent. Interestingly, among different groups, the age of Y-haplogroup R1a1* was highest in scheduled castes/tribes when compared with Central Asians and Eurasians. These observations weaken the hypothesis of introduction of this haplogroup and the origin of Indian higher most castes from Central Asian and Eurasian regions, supporting their origin within the Indian subcontinent. Further, a particular population group of northern India, the Kashmiri Pandits (KPs), showed the highest variance (0.52) and thus the respective age (Table 3). Another north Indian population group, Himachal Brahmins, also showed higher variance (0.43) than that of the average Indian population.

High frequency of Y-haplogroup R1a1* in tribal populations and ancestral Y-haplogroup R1a* in the Indian subcontinent
Y-haplogroup R1a1* has been reported to be present in the tribal population in many of the earlier studies, but with very less frequency. In this study, a tribe named Saharia from Madhya Pradesh (Central India) showed the presence of R1a1* with high diversity in 19/71 males (26.76%), negating the idea of later admixture or some founder effect. Similar observations were made in the Chenchu tribe of Andhra Pradesh,24 with a high percentage (26.82%) of R1a1*.

Figure 4. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Median joining network based on Y-STR haplotypes showing the relationship between Kashmiri and Saharia Y-chromosomes bearing Y-haplogroups R1a* and R1a1*. Biallelic markers M17 and SRY10831b were also included and given the highest weight. The root of the network represents an individual with M173-R1* (x SRY10831b-R1a).

Apart from the observation of a simultaneous presence of R1a*, the ancestral haplogroup of R1a1* was also observed in this study with a highest ever known frequency in the two population groups KPs and Saharia. Incidentally, KPs are Brahmins, whereas Saharia is a tribal population group. Scanty representation of the R1a* haplogroup and its ancestral lineages (R*, R1*) in any of the geographical regions and the presence of the R1a1* haplogroup at high frequency across Central Asia and Eurasia had kept alive the question of the origin of R1a1* and associated conflicts. With the high-resolution analyses of the haplogroup (R1) in some population groups that were absent in the earlier studies and with the addition of published datasets, we were able to provide a clearer picture of the origin of R1a1* haplogroup and solve the existing conflict in literature. The calculated age for the haplogroup R1a* in both the population groups showed fascinating results. It was observed that the variance (0.43) of R1a*, and hence the respective age of this ancestral haplogroup, was far less in Kashmiris than the observed variance (0.52) and age of the derived R1a1*. However, a variance of 0.6 was observed in the Saharia tribe for R1a*, providing the age of 21 739.13 with 95% CI 15 789.47–34 883.72 years to this haplogroup. The haplogroup R1a1* was found to have an age of 13 043.48 and 95% CI 9473.68–20 930.23 years. To resolve the contradiction in these observations, we tried to explore the whole of the R1a lineage in these two population groups. By providing higher weight to the SNP (M17 that defines R1a1*) in the median joining network of Y-STR haplotypes within the R1a lineage among KPs and Saharia (Figure 4), we were able to elucidate some important inferences based on the clustering of the haplotypes. Two main clusters differentiating R1a* and R1a1* haplogroups were observed at the first instance. Further, subclustering based on population groups could be seen within these major clusters. However, few individuals belonging to KPs were seen in the Saharia population group clusters and vice versa, representing both R1a* and R1a1* haplogroups. It was particularly interesting to observe close overlaps in R1a1* cluster. Further, the long branches and less networking in both of the clusters (R1a* and R1a1*) again indicated bottlenecks and expansions, eliminating many of the haplotypes and resulting in long branches in the median joining tree. The exclusive high presence of the ancestral R1a* lineage in KPs and Saharias, their level of sharing, observed by way of a PSA of 0.51 (based on the average of Y-STRs within R1a*) and clustering in the network, suggested their deep common ancestry, a probable source population for the origin of R1a1* and for Brahmins, which later on differed in the two population groups. This observation of a close relationship was reflected in the MDS plot based on FST values obtained from a haplotypic analysis of 6Y-STRs within R1a1* (Supplementary Figure 5). Some of the other evidences hinting at this closeness are reflected in the cultural practices as well as folklores of these population groups.

Conclusions
The observation of R1a* in high frequency for the first time in the literature, as well as analyses using different phylogenetic methods, resolved the controversy of the origin of R1a1*, supporting its origin in the Indian subcontinent. Simultaneously, the presence of R1a1* in very high frequency in Brahmins, irrespective of linguistic and geographic affiliations, suggested it as the founder haplogroup for the population. The co-presence of this haplogroup in many of the tribal populations of India, its existence in high frequency in Saharia (present study) and Chenchu tribes, the high frequency of R1a* in Kashmiri Pandits (KPs—Brahmins) as well as Saharia (tribe) and associated phylogenetic ages supported the autochthonous origin and tribal links of Indian Brahmins, confronting the concepts of recent Central Asian introduction and rank-related Eurasian contribution of the Indian caste system.

However, there is a scanty representation of Y-haplogroup R1a1 subgroups in the literature as well as in this study. The known subgroups (R1a1a, R1a1b and R1a1c), which are defined by binary markers M56, M157 or M87, respectively (Supplementary Figure 1), were not observed. In such a situation, it is likely that this haplogroup (R1a1*) is a polyphyletic (or paraphyletic) group of Y-lineages. It is, therefore, very important to discover novel Y chromosomal binary marker(s) for defining monophyletic subhaplogroup(s) belonging to Y-R1a1* with a higher resolution to confirm the present conclusion. Further, the under-representation of phylogenetic data of the population groups of North India in the literature and our observations hint at the immense need of phylogenetic explorations in the northern most Himalayan regions of India, which might have acted as an incubator of many ancient lineages, to obtain a clearer picture of the peopling of India and Eurasia.

Sharma, Swarkar, et al. "The Indian origin of paternal haplogroup R1a1* substantiates the autochthonous origin of Brahmins and the caste system." Journal of human genetics 54.1 (2009): 47-55.

Last Edit: Aug 29, 2014 4:53:45 GMT by Admin

Admin
Administrator

Posts: 72,768

Y-DNA Haplogroup R1a Apr 20, 2016 1:12:25 GMT

Quote

Post by Admin on Apr 20, 2016 1:12:25 GMT

FIGURE 1
The phylogenetic and geographic structure of Y-chromosome haplogroup R1a

Phylogeography
We measured R1a haplogroup frequency by population (Supplementary Table 4). Of the 2923 hg R1a-M420 samples, 2893 were derived for the M417/Page7 mutations (1693 non-Roma Europeans and 1200 pan-Asians), whereas the more basal subgroups were rare. We observed just 24 R1a*-M420(xSRY10831.2), 6 R1a1*-SRY10831.2(xM198), and 12 R1a1a1-M417/Page7*(xZ282,Z93). We did not observe a single instance of R1a1a-M198*(xM417,Page7), but we cannot exclude the possibility of its existence.
Of the 1693 European R1a-M417/Page7 samples, more than 96% were assigned to R1a-Z282 (Figure 2), whereas 98.4% of the 490 Central and South Asian R1a lineages belonged to hg R1a-Z93 (Figure 3), consistent with the previously proposed trend.31 Both of these haplogroups were found among Near/Middle East and Caucasus populations comprising 560 samples.

Figure 2.

Subgroups of both R1a-Z282 and R1a-Z93 exhibit geographic localization within the broad distribution zone of R1a-M417/Page7. Among R1a-Z282 subgroups (Figure 2), the highest frequencies (~20%) of paragroup R1a-Z282* chromosomes occur in northern Ukraine, Belarus, and Russia (Figure 2b). The R1a-Z284 subgroup (Figure 2c) is confined to Northwest Europe and peaks at ~20% in Norway, where the majority of R1a chromosomes (24/26) belong to this clade. We found R1a-Z284 to be extremely rare outside Scandinavia. R1a-M458 (Figure 2d) and R1a-M558 (Figure 2e) have similar distributions, with the highest frequencies observed in Central and Eastern Europe. R1a-M458 exceeds 20% in the Czech Republic, Slovakia, Poland, and Western Belarus. The lineage averages 11–15% across Russia and Ukraine and occurs at 7% or less elsewhere (Figure 2d). Unlike hg R1a-M458, the R1a-M558 clade is also common in the Volga-Uralic populations. R1a-M558 occurs at 10–33% in parts of Russia, exceeds 26% in Poland and Western Belarus, and varies between 10 and 23% in the Ukraine, whereas it drops 10-fold lower in Western Europe. In general, both R1a-M458 and R1a-M558 occur at low but informative frequencies in Balkan populations with known Slavonic heritage. The rarity of R1a-M458 and R1a-M558 among Central Asian and South Siberian R1a samples (4/301; Supplementary Table 4) suggests low levels of historic Slavic gene flow.

In the complementary R1a-Z93 haplogroup, the paragroup R1a-Z93* (Figure 3b) is most common (>30%) in the South Siberian Altai region of Russia, but it also occurs in Kyrgyzstan (6%) and in all Iranian populations (1–8%). R1a-Z2125 (Figure 3c) occurs at highest frequencies in Kyrgyzstan and in Afghan Pashtuns (>40%). We also observed it at greater than 10% frequency in other Afghan ethnic groups and in some populations in the Caucasus and Iran. Notably, R1a-M780 (Figure 3d) occurs at high frequency in South Asia: India, Pakistan, Afghanistan, and the Himalayas. The group also occurs at >3% in some Iranian populations and is present at >30% in Roma from Croatia and Hungary, consistent with previous studies reporting the presence of R1a-Z93 in Roma.31, 51 Finally, the rare R1a-M560 was only observed in four samples: two Burushaski speakers from north Pakistan, one Hazara from Afghanistan, and one Iranian Azeri.

Figure 3.

Origin of hg R1a
To infer the geographic origin of hg R1a-M420, we identified populations harboring at least one of the two most basal haplogroups and possessing high haplogroup diversity. Among the 120 populations with sample sizes of at least 50 individuals and with at least 10% occurrence of R1a, just 6 met these criteria, and 5 of these 6 populations reside in modern-day Iran. Haplogroup diversities among the six populations ranged from 0.78 to 0.86 (Supplementary Table 4). Of the 24 R1a-M420*(xSRY10831.2) chromosomes in our data set, 18 were sampled in Iran and 3 were from eastern Turkey. Similarly, five of the six observed R1a1-SRY10831.2*(xM417/Page7) chromosomes were also from Iran, with the sixth occurring in a Kabardin individual from the Caucasus. Owing to the prevalence of basal lineages and the high levels of haplogroup diversities in the region, we find a compelling case for the Middle East, possibly near present-day Iran, as the geographic origin of hg R1a.

Spatial dynamics of R1a lineage frequencies
We conducted a spatial autocorrelation analysis of the two primary subgroups of R1a (Z282 and Z93) and of each of their subgroups independently (Supplementary Figure 3). Each correlogram was statistically significant. We observed clinal distributions (continually decreasing frequency with increasing geographic distance) across a large geographic area in the two macrogroups and in M558 and M780 as well. One group (Z2125) did not reveal any discernible pattern, and the analysis of four groups (Z282*, Z284, M458, and Z93*) indicated potential clinal distributions that do not extend across the full geographic range under study. Therefore, we also analyzed partial ranges for Z282* and M458 in Europe, the Caucasus, and the Middle East, and for Z284 in Europe, but these partial range analyses also failed to yield evidence of clinal distributions.

Figure 4.

We also conducted PCA of R1a subgroups (Figure 4). The first principal component explains 21% of the variation and separates European populations at one extreme from those of South Asia at the other. The second explains 14.7% of the variation and is driven almost exclusively by the high presence of M582 among some Jewish populations, particularly the Ashkenazi Jews. PC2 separates them from all other populations. When we consider haplogroups rather than populations (Supplementary Figure 4), we see that the clustering of European populations is due to their high frequencies of M558, M458, and Z282*, whereas the M780 and Z2125* lineages account for the South Asian character of the other extreme.

Our phylogeographic data lead us to conclude that the initial episodes of R1a-M420 diversification occurred in the vicinity of Iran and Eastern Turkey, and we estimate that diversification downstream of M417/Page7 occurred ~5800 years ago. This suggests the possibility that R1a lineages accompanied demic expansions initiated during the Copper, Bronze, and Iron ages, partially replacing previous Y-chromosome strata, an interpretation consistent with albeit limited ancient DNA evidence.54, 60 However, our data do not enable us to directly ascribe the patterns of R1a geographic spread to specific prehistoric cultures or more recent demographic events. High-throughput sequencing studies of more R1a lineages will lead to further insight into the structure of the underlying tree, and ancient DNA specimens will help adjudicate the molecular clock calibration. Together these advancements will yield more refined inferences about pre-historic dispersals of peoples, their material cultures, and languages.57, 61, 62

Figure 5.

With reference to Figure 1, all fully sequenced R1a individuals share SNPs from M420 to M417. Below branch 23 in Figure 5, we see a split between Europeans, defined by Z282 (branch 22), and Asians, defined by Z93 and M746 (branch 19; Z95, which was used in the population survey, would also map to branch 19, but it falls just outside an inclusion boundary for the sequencing data4). Star-like branching near the root of the Asian subtree suggests rapid growth and dispersal. The four subhaplogroups of Z93 (branches 9-M582, 10-M560, 12-Z2125, and 17-M780, L657) constitute a multifurcation unresolved by 10 Mb of sequencing; it is likely that no further resolution of this part of the tree will be possible with current technology. Similarly, the shared European branch has just three SNPs.

We caution against ascribing findings from a contemporary phylogenetic cluster of a single genetic locus to a particular pre-historic demographic event, population migration, or cultural transformation. The R1a TMRCA estimates we report have wide confidence intervals and should be viewed as preliminary; one must sequence tens of additional R1a samples to high coverage to uncover additional informative substructure and to bolster the accuracy of the branch lengths associated with the more terminal portions of the phylogeny. Although some of the SNPs on the lineages we have defined by single SNPs are undoubtedly rare (eg, the Z2125 sub-hg M434, Figure 1, Supplementary Table 4), it remains possible that future genotyping effort using the SNPs in Supplementary Data File 1 may expose other substructure at substantial frequency, commensurate with more recent episodes of population growth and movement. In addition, high coverage sequences using multiple male pedigrees sampled across various haplogroups in the global Y phylogeny will be needed to more accurately estimate the Y-chromosome mutation rate. Nonetheless, despite the limitations of our small sample of R1a sequences, the relative shortness of the branches and their geographic distributions are consistent with a model of recent R1a diversification coincident with range expansions and population growth across Eurasia.

European Journal of Human Genetics (2015) 23, 124–131; doi:10.1038/ejhg.2014.50; published online 26 March 2014

Admin
Administrator

Posts: 72,768

Y-DNA Haplogroup R1a Jul 14, 2016 22:39:44 GMT

Quote

Post by Admin on Jul 14, 2016 22:39:44 GMT

R1a-Z282 is specific to Europe, highly concentrated in the Pontic-Caspian steppe. Z93 and Z94 make up the Asian branch of R1a which spread to Central Asia and India and they are associated with the Kurgan culture and the Indo-Aryans. R1a-Z280 is associated with the Baltic-speaking people and its descendant subclade CTS3402 is mainly found in the Dinaric Alps spanning from Albania to Serbia, especially among the Croats. R1a-CTS3402 has high frequencies in Croatia and Southern Poland and CTS3402 may be linked to the Slavic expansion to the Balkans.

The CTS3402 tribe in the Balkans are more likely to have migrated from the Baltic region or Southern Poland to the Balkans, while the Asian branch of R1a (Z93, Z94) could have reached the Balkans from Central Asia or Persia. R1a-M582 is present in 33.8% of Ashkenazi R1a male lineages. M582 is absent in Eastern Europeans but present in non-Jewish Near Easterners. 4% of Ashkenazi Jews also belong to Haplogroup R1a1a (R-M17), which is shared with Eastern European population.

Previous Y-chromosome studies have demonstrated that Ashkenazi Levites, members of a paternally inherited Jewish priestly caste, display a distinctive founder event within R1a, the most prevalent Y-chromosome haplogroup in Eastern Europe. Here we report the analysis of 16 whole R1 sequences and show that a set of 19 unique nucleotide substitutions defines the Ashkenazi R1a lineage. While our survey of one of these, M582, in 2,834 R1a samples reveals its absence in 922 Eastern Europeans, we show it is present in all sampled R1a Ashkenazi Levites, as well as in 33.8% of other R1a Ashkenazi Jewish males and 5.9% of 303 R1a Near Eastern males, where it shows considerably higher diversity. Moreover, the M582 lineage also occurs at low frequencies in non-Ashkenazi Jewish populations. In contrast to the previously suggested Eastern European origin for Ashkenazi Levites, the current data are indicative of a geographic source of the Levite founder lineage in the Near East and its likely presence among pre-Diaspora Hebrews.

The widespread Eurasian distribution zone of haplogroup R1a has been mapped for over a decade8, 10, 12, 13 since the first reports that the M17 deletion grouped Y-chromosomes into a monophyletic clade that exceeded 30% frequency in some Asian populations14 and over 50% in Eastern Europe15. Several single nucleotide polymorphisms (SNPs) phylogenetically equivalent to M17, such as M198, are now known16. Nonetheless, the internal structure of the R1a haplogroup has long resisted fractionation until the recent separation of its post-glacial European and Asian coancestry was partially achieved with the discovery of the M458 SNP16. Efforts to clarify the genetic ancestry of East European Jews have been ongoing17. Progress towards disentangling the internal structure of haplogroup R1a is integral to our ability to resolve the debate over the geographic origin of the Ashkenazi Jewish Levite founding lineage, which is defined by a short tandem repeat (STR)-based haplotype that is rare elsewhere18.

Figure 1: Y-chromosome haplogroup R1 phylogeny and haplogroup R1a-M582 STR haplotypes diversity.

Haplogroup R1 whole Y-chromosome phylogeny
We have determined at high coverage the sequence of eight Jewish and five non-Jewish Y-chromosomes (Supplementary Table S1) belonging to haplogroup R1 (Fig. 1a and Supplementary Fig. S1). We reconstructed the phylogenetic relationships of the newly sequenced samples in the context of three other available high coverage sequences from this haplogroup using binary SNP data (Supplementary Data 1) from a total of 8.97 Mbp of sequence from the non-recombining region of nine Y-chromosomes28. For both sub-clades of R1, most Jewish sequences cluster separately from the clades that in previous studies have been shown to be common in Europe: in R1b where more than 90% of European lineages are within the L11/L52 clade29, three of the newly sequenced Jewish Y-chromosomes cluster within its copanion clade defined by the Z2105 mutation; in R1a where the majority of European individuals belong to the S198 (ref. 30) and M458 branches16, all four newly sequenced Jewish Y-chromosomes fall into the clade defined by Z94 mutation. Furthermore, two Ashkenazi R1a chromosomes were found to share 19 SNPs (Supplementary Fig. S1) of which six were found to be shared with an Iberian individual from the 1,000 Genomes database. One of the mutations defining this clade including Ashkenazi and Iberian samples, which we designate M582 (AKA CTS2253), was subjected to genotyping in 2,834 R1a samples to determine its geographic distribution, diversity and divergence times.

Population-based survey for R1a-M582
Among non-Jewish populations, the overall frequency of R1a-M582 was found to be 0.15% (22/15,138) and among R1a-M198 it was 0.81% (22/2,711) (Table 1). The geographic distribution of the haplogroup appears to be limited to West Eurasia, as we did not observe it in South Asia, Central Asia or Southern Siberia. Haplogroup R1a-M582 was only sporadically observed in Europe, the Diaspora residence of Ashkenazi Jews. Notably, it was not identified among 2,149 samples (including 922 R1a-M198) of non-Jews from East Europe, where the Ashkenazi Jewish community flourished in recent centuries (Table 1). While the frequency of R1a in the paternal gene pool of Eastern European Slavonic and non-Slavonic peoples16, 18, 31 reaches as high or higher than 50%, the deeper phylogenetic resolution reported here renders the presumption that similarly high frequencies of R1a among Ashkenazi Levites reflect substantial gene flow or a founder effect from their hosts highly unlikely. Within 1,068 West/North European samples (106 R1a-M198), M582 was observed in just one German sample, and among 3,756 Central/South European samples (710 R1a-M198), it was found only in one Hungarian and one Slovakian sample (Table 1). It is important to note that many of the samples from individuals of European descent from Germany, The Netherlands, England, Ireland and Norway were collected in USA without ethnicity or religious affiliation information, so some of them might represent contemporary Ashkenazi Jews residing in the United States.

As such, our results are likely more conservative because fewer of our R1a-M582 samples actually have non-Jewish European ancestry. Among 3,739 Near Eastern samples (303 R1a-M198), R1a-M582 was identified in various populations, with the highest frequency occurring within Iranians collected from the southeastern Kerman population who self-identified as Persians, northwestern Iranian Azeri and in Cilician Anatolian Kurds, at 2.86%, 2.50% and 2.83%, respectively (Table 1). In contrast, among 2,164 samples from the Caucasus (211 R1a-M198), R1a-M582 was found in just one Nogay sample (Table 1). As the overall frequencies obtained for R1a-M582 were low, we applied Fisher's exact test to verify that the different sample sizes did not bias our results (Supplementary Table S2). Out of a total of eight tests for the entire population, all but one was significant at the nominal 0.05 level, and all but three were still significant after a Bonferroni correction. For the R1a-M198 samples only, all eight tests were significant at the 0.05 level and all but one significant after Bonferroni correction.

Of the 821 non-Ashkenazi Jewish samples, 36 (4.38%) belonged to haplogroup R1a, and ten (1.22%) belonged to R1a-M582 (Table 2). These frequencies are similar to R1a-M582 frequencies observed in non-Jewish Near Eastern populations. Of the 10 non-Ashkenazi R1a-M582 individuals, two were from the North African Algerian Jewish community, six belonged to Spanish expulsion descendent communities of Slovenia, Turkey and Bulgaria, and two were from individuals reporting their last known parental origin as Israel. Significantly, all eight individuals for whom caste information was available self-identified as Levites. Caste information was unavailable for the two Bulgarian individuals (Supplementary Data 2). Moreover, out of nine R1a-M198 non-Ashkenazi Levites, eight were R1a-M582. Among the 600 Ashkenazi Jews, 87 belonged to haplogroup R1a-M198, and of these, 80 (91.95%) were confirmed to belong to haplogroup R1a-M582 (Table 2). The Ashkenazi samples included a total of 279 Israelites, 110 Cohens, 97 Levites and 114 unclassified samples in which the counts and frequencies of haplogroup R1a-M198/R1a-M582 were 12 (4.3%)/7 (2.5%), 1 (0.9%)/1 (0.9%), 63 (64.9%)/63 (64.9%) and 11 (9.6%)/9 (7.9%), respectively (Table 2). Remarkably, all Ashkenazi Levite R1a samples belonged to haplogroup R1a-M582, and this haplogroup is virtually absent among members of the Cohen caste.

The direction of gene flow involving the Ashkenazi Levite R1a-M582 lineage warrants discussion. The presence of haplogroup R1a-M582 at minute frequencies among Europeans could theoretically suggest introgression into the Ashkenazi Levite community and subsequent expansion among Ashkenazi Levites. Alternatively, the European samples can represent Ashkenazi admixture into the general European gene pool due to assimilation that has occurred in Western Europe throughout the centuries. While it is impossible to refute the former, a few arguments render the latter more likely. The European samples carrying haplogroup R1a-M582 haplotypes add no further diversity to that found among Ashkenazi Jews, despite having been identified at different European locations. These haplotypes are actually identical or one step removed from the Levite modal haplotype. If haplogroup R1a-M582 were of general ancient European ancestry, we would have observed at least some baseline diversity and genetic distance expressed in a few STR-based mutational events between the lineages evolving in the general European gene pool and the one lineage whose evolution is confined to the Ashkenazi samples. While this pattern could not be demonstrated among Europeans, it was clearly evident among Near Eastern samples.

Near Eastern populations are the only populations in which haplogroup R1a-M582 was found at significant frequencies (Table 1). Moreover, the representative samples displayed substantial diversity even within this geographic region (Fig. 1b). Higher frequencies and diversities often suggest lineage autochthony. Hence, we can assess whether or not the origin of haplogroup R1a-M582 is in present-day Iran and eastern Anatolia, or rather the broader region of the Near East. Our data demonstrate the occurrence of R1a-M582 among different Iranian populations, among Kurds from Cilician Anatolia and Kazakhstan, and among Ashkenazi and non-Ashkenazi Jews. These observations, and the STR network delineating an internal R1a-M582 structure, might attest to a broad Near Eastern distribution range of this minor haplogroup that survived to the present day at low frequencies among Iranian Kerman, Iranian Azeri, Kurds and Jews. Haplogroup R1a-M582 was not detected in samples from Iraq or among Bedouins, Druze and Palestinians sampled in Israel.

Another notable observation regards the distribution of haplogroup R1a-M582 among Ashkenazi and non-Ashkenazi Levites. While R1a-M582 occurs at 64.9% (63/97) among Ashkenazi Levites, it comprises just 15.7% (8/51) of the non-Ashkenazi Levite paternal gene pool (Table 2). Among non-Ashkenazi Jews, R1a-M582 was observed only in Levites, and the observed sub-haplogroup shares the same STR signature as that seen in Ashkenazi Levites. A few demographic scenarios can account for this observation. Clearly, a joint Levantine origin before the Diaspora split into the Ashkenazi and non-Ashkenazi Jews must be considered. Under this scenario, this particular R1a-M582 Levite lineage existed among the ancient Hebrews and was carried to the various Jewish Diaspora communities in a manner similar to that of the Cohen Modal Haplotype21, 22, 23. However, recent studies of genome-wide diversity point to recent shared Levant ancestries and a close genetic proximity among members of Ashkenazi, North African and Spanish Exile Jewish communities17, 36, 37. Similarly, some of the mitochondrial DNA Ashkenazi founding lineages are also found among Spanish Exiles20. Therefore, another possible scenario is that of continuous gene flow between Ashkenazi, North African and Spanish Exile Jewish communities. Under this scenario, Ashkenazi Levites must have repeatedly and episodically introgressed into non-Ashkenazi communities, maintaining their Levite status while abandoning their Ashkenazi affiliation.

Nature Communications 4, Article number: 2928 doi:10.1038/ncomms3928

Last Edit: Jul 14, 2016 22:40:38 GMT by Admin

Admin
Administrator

Posts: 72,768

Y-DNA Haplogroup R1a Jul 15, 2016 22:44:56 GMT

Quote

Post by Admin on Jul 15, 2016 22:44:56 GMT

Tuvan Children--Brother and Sister--from Ger Camp near Russian Border in NW Mongolia

In recent years, indigenous ethnic groups of Altai–
Sayan region were among the most actively studied by
the specialists in different fields of human population
genetics and medical genetics. Considerable interest
to the problem of identifying the gene pool structure of
South Siberian populations is associated with the
complexity of their ethnogeny. According to anthro
pological data, metisation was the leading factor of
population differentiation on the territory of steppe
and forest steppe parts of South Siberia [1, 2]. The
processes of the assimilation and fusion with the par
ticipation of Mongoloid and Caucasoid population
groups played a major role in the formation of modern
Turkicspeaking populations of South Siberia, includ
ing Tuvinians. In the Neolithic, Bronze, and Early
Iron Ages the territory of Tuva was the part of the hab
itat of ancient Caucasoid population. Later, on this
territory, the cultures of Scythian–Siberian were
developed [3]. The penetration of Central Asian Mon
goloid component to the territory of South Siberia can
be dated back to the 7th–6th age BP. The emergence
of forest, taiga Mongoloid component is also attrib
uted to about this time [1, 4]. In the course of time,
gradual increase of the Mongoloid component, from
the prevalence of Caucasoid component in Scythian
time to the formation of modern Central Asian
anthropological type of Tuvinians in 13th–14th cen
tury AD was observed [3–5]. Genetic relationships
between different population groups on the territory of
Siberia caused by considerable migrations, which
especially intensified during Bronze Age, Early and
Late Iron Age, and Middle Ages, resulted in the for
mation of transitional Mongoloid–Caucasoid popu
lations in the Altai–Sayan region.

In Tuvinians, the most frequent Ychromosome
variant was haplogroup N1b, which constituted 24%
in the total gene pool. This lineage was the most abun
dant in the groups from the center and south of the
republic. Moreover, in the latter group, almost 39% of
all Y chromosomes belonged to this lineage. In this
respect, it should be noted that no other haplogroup
with such a high frequency was detected in any popu
lation. It seems likely that high frequency of this hap
logroup in modern Tuvinians reflects the contribution
the Samoyedic ethnic groups that previously popu
lated the territory of Tuva to their gene pool.
Haplogroup N1c1 is the second highly frequent
haplogroup in Tuvinians (19% out of all samples). In
the total sample, the frequency of this haplogroup is
only 5% lower than that of haplogroup N1b. More
over, in the west of Tuva, Nic1 was present in about
30% of the samples examined. The distinguishing fea
ture of the populations from the west and northeast of
the republic is that the frequency of haplogroup N1c1
in these populations is higher than that of N1b. At the
same time, in other parts of Tuva, the frequency of
haplogroup N1b is orders of magnitude higher than
that of N1c1. The lowest frequency of haplogroup
N1c1 was observed in the south and southeast of Tuva.
Thus, haplogroups of N group displayed a gradient of
decreasing frequencies, from west to east (for haplo
group N1b) and northwest to southeast (for haplo
group N1c1).

Haplogroup Q1a3 was found in 14% of all Tuvinian
specimens. This haplogroup was detected with the
highest frequency in the eastern samples (25%), while
it was absent in the samples from the south of the
republic. Q1a3 demonstrated the east–west gradient
of decreasing frequencies. The highest for Tuvinians
frequency of haplogroup Q1a3, observed in mountain
taiga and southeastern regions, as well as in Todja,
probably results from geographic inaccessibility of
these regions and, as a consequence, from relative
genetic isolation of local populations. Under these
conditions, in the gene pool of the Tuva population, the
highest proportion of ancient component was preserved.
Alternatively, this situation can be associated with the fact
that the gene pool of Todja Tuvinians was formed with
more active participation of Kets, who are characterized
by the high frequency of haplogroup Q1a3.

The proportion of haplogroup R1a1a in the total
gene pool of Tuvinians is somewhat lower (12%). This
lineage is characterized by high frequencies in the
populations of South Siberia. Furthermore, R1a1a
dominates in Altaians, constituting 60% [10]. In Kha
kasses, this haplogroup is the second most frequent
(28%) [13]. Interestingly, in western Truvinian sample,
the frequency of haplogroup R1a1a was considerably
lower than in the central sample. Based on the close
ness of the Altai, which is populated by the represen
tatives of a more Caucasoid SouthSiberian racial
type, it would be reasonable to expect the west–east
decrease of the R1a1a frequency on the territory of
Tuva. However, this was not observed, and the change
of the haplogroup frequency was rather the opposite,
as the eastern samples demonstrated maximum fre
quency of this haplogroup. At the first glance, the
result obtained is paradoxical. Specifically, in terms of
anthropology, the most Caucasoid population of the
western parts of Tuva displays the minimum of haplo
group R1a1a, while in the most Mongoloid population
of Todja, the maximum of this haplogroup is observed.

Archeological evidence indicates a Scythian presence possibly as early as the 9th century BC in eastern Tuva and the eastern Truvinians are descended from the Scythians who had R1a1a. Haplogroup R1a1a reaches the maximum frequency (26.1%) in the northeastern part of the Tuva Republic because of their Scythian ancestry, while those who live in the western part (8.3%) have less Scythian ancestry and more Mongolian admixture. A site of early Scythian kurgan burials is located in the Tuva Republic, some 60 kilometers the north-west of Kyzyl, and the western half of Tuva was repeatedly invaded by the Mongols.

ISSN 10227954, Russian Journal of Genetics, 2013, Vol. 49, No. 12, pp. 1236–1244.

Last Edit: Apr 12, 2018 20:28:23 GMT by Admin

Admin
Administrator

Posts: 72,768

Y-DNA Haplogroup R1a Jul 17, 2016 22:38:34 GMT

Quote

Post by Admin on Jul 17, 2016 22:38:34 GMT

Haplogroup R1a makes up 51% among the Pashtun people but 16-18% of them also belong to Haplogroup Q, which is typically found in Mongolia. The Hazara and Pashtun tribes in Afghanistan have a mixture of western Eurasian and eastern Eurasian haplogroups because of their patrimonial and maternal relations to Mongol people. The frequency of the Asian haplogroup C in the Hazara is 40% and they cluster with the Pashtuns and other Afghan tribes in the PCA, plotted somewhere between East Asians and Caucasians. The prevailing Y-chromosome lineage in the Pashtuns is R1a1a-M17, which has the highest diversity among populations of the Indus Valley, and the ancestors of the Pashtun people originally migrated from India and then admixed with local Asians in Afghanistan.

Genotyping revealed 32 halpogroups present in Afghanistan's ethnic groups among our samples. Haplogroups R1a1a-M17, C3-M217, J2-M172, and L-M20 were the most frequent when Afghan ethnic groups were pooled, together comprising >66% of the chromosomes. Absolute and relative haplogroup frequencies are tabulated in Table S4.

Haplogroup frequencies across the major ethnic groups revealed large differences. In particular, frequencies of haplogroup C3-M217, which is mainly found in East Asia, and haplogroup R1a1a-M17, which is found in Eurasia, varied substantially among the Afghan groups. C3-M217 was significantly more frequent (p = 4.55×10−9) in Uzbeks (41.18%) and Hazaras (33.33%) than it was in Tajiks (3.57%) and Pashtuns (2.04%). On the other hand, R1a1a-M17 was significantly more frequent (p = 3.00×10−6) in Pashtuns (51.02%) and Tajiks (30.36%) than in Uzbeks (17.65%) and Hazaras (6.67%). RM networks of C3-M217 (Figure S1A) and R1a1a-M17 (Figure S1B) show that when a haplogroup was infrequent in an ethnic group, its haplotypes existed on branches not shared with other Afghans, suggesting that the underrepresented haplogroups are not the result of a gene flow between the ethnic groups, but probably a direct assimilation from source populations.

Figure 1. PCA derived from Y-chromosomal haplogroup frequencies.

Haplogroups autochthonous to India [15]; L-M20, H-M69, and R2a-M124 were found more (p = 0.004) in Pashtuns (20.41%) and Tajiks (19.64%) than in Uzbeks (5.88%) and Hazaras (5%). E1b1b1-M35 was found in Hazaras (5%) and Uzbeks (5.88%) but not in Pashtuns and Tajiks. RM network of E1b1b1-M35 (Figure S1C) shows that Afghanistan's lineages are correlated with Middle Easterners and Iranians. We also note the presence of the African B-M60 only in Hazara, with a relatively recent common founder ancestor from East Africa as shown in the RM network (Figure S1D).

PCA of the haplogroups frequency (Figure 1) also shows differences among Afghans. Although the worldwide populations are mostly clustered according to geography, Afghan groups appear to show more affinity to non-Afghans than to each others. Pashtun and Hazara in Afghanistan and Pakistan show affinity to their ethnic groups across borders. The Afghan Tajiks show equal distance to Central Asia and to Iran/Caucasus/West Russia. The Afghan Hazara, Afghan Uzbek, and Pakistan Hazara sit between East Asia and the Middle East/Europe-Caucasus/West Russia cluster.

More details about the structure of the Afghan population appear in the MDS of the 's (Figure 2B) which shows that the Afghan Pashtun and Tajik are closer to North and West Indians than to the other Afghans; Hazara and Uzbek. This cluster also sits between East Europeans and Iranians more close to the Iranians especially to East Azerbaijan. Furthermore, Barrier (Figure 2A) shows that Barrier IV splits the Afghan populations separating the Hazara and Uzbek from the Pashtun, Tajik and the Indian populations, creating groups of populations that have less variation within the groups (2.30%, p<0.001) and more variation among groups (10.48%, p<0.001) compared to populations grouped by region or country (within groups = 4.95%, p<0.001, among groups = 7.16%, p<0.001) (Table S5).

Figure 2. Population genetic structure vs geography.

To explore the time depth in which the above reported structures have emerged, we employed BATWING to create hypotheses on historical population splitting and coalescent events, reflecting dominating genetic ancestral structures identified in BATWING's modal trees from which the current populations have emerged (Table S6). The BATWING results showed that most of the regional splits occurred around 10 kya (95% CI 7,100–15,825) (Figure 3). These splits coincide with post LGM expansions that have led to the Neolithic agricultural revolution. During this period Afghans, Iranians, Indians and East Europeans most likely emerged as distinct unstructured populations. BATWING showed another wave of splits that started later and may have created the inter-population structures. This second wave of splits started in Afghans 4.7 kya (95% CI 2,775–7,725), marking the start of civilization building and displacements, and these splits appear to have continued to nearly modern times. BATWING results in general corroborated the geographical splits identified by BARRIER.

Figure 3. Composite BATWING population splitting.

The prevailing Y-chromosome lineage in Pashtun and Tajik (R1a1a-M17), has the highest observed diversity among populations of the Indus Valley [46]. R1a1a-M17 diversity declines toward the Pontic-Caspian steppe where the mid-Holocene R1a1a7-M458 sublineage is dominant [46]. R1a1a7-M458 was absent in Afghanistan, suggesting that R1a1a-M17 does not support, as previously thought [47], expansions from the Pontic Steppe [3], bringing the Indo-European languages to Central Asia and India.

MDS and Barrier analysis have identified a significant affinity between Pashtun, Tajik, North Indian, and West Indian populations, creating an Afghan-Indian population structure that excludes the Hazaras, Uzbeks, and the South Indian Dravidian speakers. In addition, gene flow to Afghanistan from India marked by Indian lineages, L-M20, H-M69, and R2a-M124, also seems to mostly involve Pashtuns and Tajiks. This genetic affinity and gene flow suggests interactions that could have existed since at least the establishment of the region's first civilizations at the Indus Valley and the Bactria-Margiana Archaeological Complex.

Reduced median networks. (A) C-M130, (B) R1a1a-M17, (C) E1b1b1-M35, and (D) B-M60 showing STR haplotype distributions among populations; area is proportional to haplotype frequency, and color indicates populations. Connecting lines represent putative phylogenetic relationships between haplotypes.

The E1b1b1-M35 lineages in some Pakistani Pashtun were previously traced to a Greek origin brought by Alexander's invasions [48]. However, RM network of E1b1b1-M35 found that Afghanistan's lineages are correlated with Middle Easterners and Iranians but not with populations from the Balkans.

The Islamic invasion in the 7th century CE left an immense cultural impact on the region, with reports of Arabs settling in Afghanistan and mixing with the local population [49]. However the genetic signal of this expansion is not clearly evident: some Middle Eastern lineages such as E1b1b1-M35 are present in Afghanistan, but the most prevalent lineage among Arabs (J1-M267) was only found in one Afghan subject. In addition, the three Afghans that identified their ethnicity as Arab, had lineages autochthonous to India.

Haber M, Platt DE, Ashrafian Bonab M, Youhanna SC, Soria-Hernanz DF, Martínez-Cruz B, et al. (2012) Afghanistan's Ethnic Groups Share a Y-Chromosomal Heritage Structured by Historical Events. PLoS ONE 7(3): e34288.