|
|
Post by Admin on Oct 13, 2022 7:32:28 GMT
Results
We generated double- and single-stranded next-generation sequencing libraries (45, 46) both using partial uracil-DNA glycosylase (UDG) treatment and without such treatment for two samples and enriched them for ∼1.2 million single-nucleotide polymorphisms (SNPs) (47, 48) (Dataset S1). Filtering for standard quality criteria and using a cutoff of at least 10,000 covered SNPs, we retain a dataset of 16 individuals buried in mass graves associated with the 480 BCE battle (SI Appendix, Figs. S1–S6), 5 associated with the 409 BCE battle (SI Appendix, Fig. S7), 11 attributed to the civilian population buried in Himera’s West necropolis, and 1 attributed to the civilian population buried at the East necropolis (Dataset S2). All of the soldiers are genetically sexed as males, while seven civilian individuals are genetically sexed as male and the other five as female. A total of 19 individuals were analyzed from the Iron Age site of Polizzello (ninth and eighth centuries BCE) and 2 from the site of Monte Falcone at Baucina (sixth to fifth centuries BCE), representing both males and females. One pair of second-degree relatives was detected at Polizzello (SI Appendix, Fig. S8). Overall, the average coverage on samples from the new dataset ranges from 0.011- to 4.298-fold on targeted SNPs on the autosomes with between 11,393 and 905,741 SNPs covered. Nuclear contamination point estimates on the X chromosome, which can be estimated for males, ranged from 0 to 1.6%. To contextualize the newly reported ancient individuals, we also generated new genotyping data for 96 modern-day Italian, Greek, and Cretan individuals on the Affymetrix Human Origins (HO) SNP set (49) (SI Appendix, Fig. S9 and Dataset S3).
We computed principal components (PCs) on the newly generated HO data, together with 985 previously published present-day individuals from 64 diverse West Eurasian and North African populations (49), and projected new and published ancient individuals (Datasets S3 and S4) onto the first two PCs (Fig. 2A). As previously reported (49), modern Western Eurasians form two parallel clines separated on the first PC, which distinguishes European populations from populations of the Near East and Caucasus, while the second PC roughly reflects the north–south geographical distribution of the populations. Modern Aegean and Balkan populations occupy the space between the two parallel clines, forming a genetic and geographic bridge. We ran unsupervised ADMIXTURE (50) on a worldwide set of 1,900 modern-day individuals and 2,453 ancient individuals, including our newly reported data for 2–15 clusters k and display the results for k = 6, the lowest k to differentiate genetic clusters associated with three key ancestry components for Europeans, that is, Western hunter-gatherers (WHGs), early European farmers (EEFs), and Caucasus hunter-gatherers (CHGs) (Fig. 2B and SI Appendix, Fig. S10). We observe a separation of most of the individuals into distinct clusters in PC space that remain stable considering the 95% CI ellipses of low coverage samples and when constructing PCs on a worldwide and Eurasian set of modern-day individuals (SI Appendix, Figs. S11 and S12) and that correlate with the proportion of different components in ADMIXTURE analysis. We confirm these genetic clusters with pairwise qpWave analyses (SI Appendix, Fig. S13 and Tables S2–S7), group the individuals according to their genetic cluster and chronological and archaeological contexts, and perform subsequent ancestry modeling on a per-group and per-individual basis (SI Appendix, Tables S8–S17 and Dataset S5).
|
|
|
|
Post by Admin on Oct 14, 2022 7:08:51 GMT
Fig. 2.  PCA and ADMIXTURE results for Sicilian individuals and relevant ancient samples from the literature. (A) Ancient data for individuals with data at more than 10,000 HO SNPs projected onto the first two PCs constructed with 1,081 modern-day west Eurasian and North African individuals, represented by small gray circles (SI Appendix, Fig. S8 includes population labels). Color and symbol indicate geographic origin and cultural affiliation of ancient samples, respectively. Circled symbols indicate individuals with strontium isotope signatures inconsistent with an origin in Sicily. Symbols without a black outline denote samples covered on fewer than 50,000 1240K SNPs. (B) Results of unsupervised ADMIXTURE, showing k = 6 for new Sicilian data and select ancient samples. Samples covered on fewer than 50,000 1240K SNPs shown at lower opacity. Full results are shown in SI Appendix, Fig. S1. |
|
|
|
Post by Admin on Oct 17, 2022 2:28:30 GMT
Results
We generated double- and single-stranded next-generation sequencing libraries (45, 46) both using partial uracil-DNA glycosylase (UDG) treatment and without such treatment for two samples and enriched them for ∼1.2 million single-nucleotide polymorphisms (SNPs) (47, 48) (Dataset S1). Filtering for standard quality criteria and using a cutoff of at least 10,000 covered SNPs, we retain a dataset of 16 individuals buried in mass graves associated with the 480 BCE battle (SI Appendix, Figs. S1–S6), 5 associated with the 409 BCE battle (SI Appendix, Fig. S7), 11 attributed to the civilian population buried in Himera’s West necropolis, and 1 attributed to the civilian population buried at the East necropolis (Dataset S2). All of the soldiers are genetically sexed as males, while seven civilian individuals are genetically sexed as male and the other five as female. A total of 19 individuals were analyzed from the Iron Age site of Polizzello (ninth and eighth centuries BCE) and 2 from the site of Monte Falcone at Baucina (sixth to fifth centuries BCE), representing both males and females. One pair of second-degree relatives was detected at Polizzello (SI Appendix, Fig. S8). Overall, the average coverage on samples from the new dataset ranges from 0.011- to 4.298-fold on targeted SNPs on the autosomes with between 11,393 and 905,741 SNPs covered. Nuclear contamination point estimates on the X chromosome, which can be estimated for males, ranged from 0 to 1.6%. To contextualize the newly reported ancient individuals, we also generated new genotyping data for 96 modern-day Italian, Greek, and Cretan individuals on the Affymetrix Human Origins (HO) SNP set (49) (SI Appendix, Fig. S9 and Dataset S3).
We computed principal components (PCs) on the newly generated HO data, together with 985 previously published present-day individuals from 64 diverse West Eurasian and North African populations (49), and projected new and published ancient individuals (Datasets S3 and S4) onto the first two PCs (Fig. 2A). As previously reported (49), modern Western Eurasians form two parallel clines separated on the first PC, which distinguishes European populations from populations of the Near East and Caucasus, while the second PC roughly reflects the north–south geographical distribution of the populations. Modern Aegean and Balkan populations occupy the space between the two parallel clines, forming a genetic and geographic bridge. We ran unsupervised ADMIXTURE (50) on a worldwide set of 1,900 modern-day individuals and 2,453 ancient individuals, including our newly reported data for 2–15 clusters k and display the results for k = 6, the lowest k to differentiate genetic clusters associated with three key ancestry components for Europeans, that is, Western hunter-gatherers (WHGs), early European farmers (EEFs), and Caucasus hunter-gatherers (CHGs) (Fig. 2B and SI Appendix, Fig. S10). We observe a separation of most of the individuals into distinct clusters in PC space that remain stable considering the 95% CI ellipses of low coverage samples and when constructing PCs on a worldwide and Eurasian set of modern-day individuals (SI Appendix, Figs. S11 and S12) and that correlate with the proportion of different components in ADMIXTURE analysis. We confirm these genetic clusters with pairwise qpWave analyses (SI Appendix, Fig. S13 and Tables S2–S7), group the individuals according to their genetic cluster and chronological and archaeological contexts, and perform subsequent ancestry modeling on a per-group and per-individual basis (SI Appendix, Tables S8–S17 and Dataset S5).
|
|
|
|
Post by Admin on Oct 18, 2022 6:32:23 GMT
Fig. 2. PCA and ADMIXTURE results for Sicilian individuals and relevant ancient samples from the literature. (A) Ancient data for individuals with data at more than 10,000 HO SNPs projected onto the first two PCs constructed with 1,081 modern-day west Eurasian and North African individuals, represented by small gray circles (SI Appendix, Fig. S8 includes population labels). Color and symbol indicate geographic origin and cultural affiliation of ancient samples, respectively. Circled symbols indicate individuals with strontium isotope signatures inconsistent with an origin in Sicily. Symbols without a black outline denote samples covered on fewer than 50,000 1240K SNPs. (B) Results of unsupervised ADMIXTURE, showing k = 6 for new Sicilian data and select ancient samples. Samples covered on fewer than 50,000 1240K SNPs shown at lower opacity. Full results are shown in SI Appendix, Fig. S1. |
|
|
|
Post by Admin on Oct 19, 2022 6:40:39 GMT
The Civilian Population of Himera and Surrounding Regions.
We find that the IA Sicilians (Sicily_IA), associated with the Sicani culture, form a homogenous cluster distinct from most of the individuals excavated at Himera. They fall in the PC space occupied by earlier Middle and Late Bronze Age Sicilians (Sicily_MBA and Sicily_LBA; Fig. 2A) and display similar and homogenous proportions of the genetic components maximized in WHGs, CHGs, and EEFs in the ADMIXTURE analysis (blue, green, and orange in Fig. 2B, respectively). Using qpAdm, this group can be modeled as an admixture of four sources that distantly contributed to the genetic composition of Europeans (P = 0.179): Northwestern Anatolian Neolithic farmers (Turkey_Barcin_N; 76.4 ± 1.2%), WHGs (6.4 ± 1.0%), early farmers from Iran (Iran_GanjDareh_N; 6.3 ± 1.5%), and Early Bronze Age (EBA) Steppe herders associated with the Yamnaya cultural complex (Russia_Samara_EBA_Yamnaya; 10.9 ± 1.6%), which indicates an increase of Iranian-related admixture compared with the preceding LBA Sicily group, which can be best modeled without that component (SI Appendix, Table S8). Models using more proximal Bronze Age (BA) to Iron Age (IA) ancestry sources are not rejected for two-way mixtures between a local Sicilian source (Sicily_LBA, Sicily_LBA_I10371, or Sicily_MBA) on the one hand (57.1–98.7%) and Italy_IA_Republic.SG, Spain_BA, Spain_IA, Armenia_LBA.SG, or Balkan_IA (1.3–42.9%) on the other hand (SI Appendix, Table S9). This indicates that the population might not have been completely continuous to the BA inhabitants of Sicily sampled to date and instead received some gene flow from other populations that are most plausibly from outside of Sicily. We caution that there are only published data from 15 individuals from BA Sicily, leaving the possibility that some unsampled Sicilian populations were genetically more continuous with later groups. The potential connection to Spain_IA is intriguing, as structural similarities in EBA fortifications have been documented at Castelluccio near Syracuse and artifacts, such as Bell Beakers, have been found at BA sites near Palermo (51). The arrival of amber on Sicily from Iberia has been documented to at least the third century BCE (52), if not earlier (53). Additionally, laterally spiked axes have been discovered on Sicily that likely derive from Iberian metallurgical traditions, suggesting wide Mediterranean trade networks (54). However, Leighton (1999) notes that, although “it has often been supposed that Sicilian beakers derive from Spanish types…there is no sure evidence that they represent actual imports, and since North Italian finds have multiplied in recent years and a central European origin for the form now seems likely, the case for a direct Iberian link is not as compelling as it was” (9). Ancient historians, who plausibly had access to sources that have since been lost, also discussed the possibility of connections to Iberia: Thucydides wrote of the Sicani that they were not indigenous to the island but settlers from Iberia (Thuc. 6.2.2) (55), but Diodorus Siculus disagreed with this claim and instead name them as the original inhabitants of Sicily (Diod. 5.2.4). Of course, the accounts of ancient historians writing hundreds or even thousands of years after the events they describe should not be taken as direct evidence but only as a representation of even older oral history, which is important to consider but may not be based on fact at all. Our results offer tentative support to both scenarios; however, it is unlikely that the Sicani were descended only from Iberian or other non-Sicilian populations. One line of evidence for some local genetic continuity is the almost exclusive presence of Y-chromosomal haplogroup G-Z1903 and its derivates among the males (Dataset S6), a lineage already found among Sicilian inhabitants in the Middle Bronze Age (MBA) and LBA (12) and otherwise unreported from any pre- or early historic contexts, including Iberia. One male belongs to haplogroup R-FT40455, which is a specific subtype of haplogroup R-DF27, which could reflect an Iberian source of ancestry, as it is much more common in Iberia from the BA onward than in any other region (19, 56). R-DF27 has also been observed in individuals of the EBA in Sicily (12) and thus could have that persisted to the time of the Sicani culture. We find overall low levels of runs of homozygosity among the Sicily_IA and preceding BA individuals (SI Appendix, Fig. S14 and Table S18), providing no evidence for endogamous practices and pointing toward large effective population sizes (57).
We analyzed 11 individuals, believed to represent civilians, from single graves in Himera’s West necropolis, which was in use during the sixth to fifth centuries BCE and potentially spanning the entire occupation of the site, as analyses of burial inclusions to identify temporal information for individuals at Himera is still ongoing. Only three individuals have sufficient coverage (at least 50,000 SNPs covered) to perform formal statistics, but all individuals are consistent with a broadly Central or Eastern Mediterranean genetic profile in PC analysis (PCA) and ADMIXTURE (Fig. 2 and SI Appendix, Figs. S9–S12). Of the three higher-coverage individuals, both I20166/W3182 and I20168/W3702 are cladal with Greek_LBA (P = 3.55 × 10−2 and 7.80 × 10−2, respectively). I20163/W1838 appears distinct, with models of ancestry with the highest P values (P > 0.1) involving a contribution of 38–55% of his ancestry from a local Sicilian MBA, LBA, or IA source and around 45–62% from a group closely related to Punic individuals from Sardinia that harbor North African ancestry or alternatively around 87.6 ± 3.1% Sicily_IA ancestry with 12.4± 3.1% ancestry deriving from a group represented by genetic outlier from Chalcolithic Sardinia carrying fully North African ancestry (SI Appendix, Table S14). It is notable that four of the eight individuals that could be analyzed with ADMIXTURE carry a small proportion (2.2–3.8%) of a genetic component maximized in Africans that is not found to the same extent in our other Sicilian individuals but appears in Punic and Levantine individuals (Fig. 2B), and the same individuals are shifted toward Levantine groups on the worldwide PCA (SI Appendix, Fig. S11B). Considering the low coverage of the civilian individuals, comparisons of within-group variance and centroid distance on the first two PCs nevertheless suggest that the civilian sample overlapped with the soldiers of both battles in the groups Sicily_Himera_480BCE_1 and Sicily_Himera_409BCE, described below (P = 0.218, one-way PERMANOVA (58) of centroid distance). The Himera civilian individuals were, however, genetically significantly more diverse (P = 2.51–05, t test on Euclidean distance to the group centroid) than the soldiers, who derived most of their ancestry from Aegean-related populations and plausibly represented Greek settlers or their descendants. This greater genetic diversity is likely not simply explained by the lower coverage of the civilian individuals (SI Appendix, Figs. S12 and S15). A greater sample size ideally from a wider diversity of burial contexts at Himera would be necessary to test whether they could represent a genetically stratified subset of a more-diverse civilian population. Greek colonies were the meeting grounds of culturally and genetically diverse people (16, 17, 59). Those living in Greek poleis and apoikiai (independent cities and colonies, respectively) included people with ancestors from the Greek mainland and Greek isles but also people from the colony’s hinterlands and members of other cultural groups, such as Phoenicians and Etruscans, which could explain the signal we observe (2, 59–61). Himera, especially, may have had strong Phoenician ties due to its proximity to neighboring Punic settlements and the Himeran tyrant Terillus’ alliance with Carthage (Hdt. 7.165).
The individual buried at Himera’s East necropolis, which was in use from the seventh to fifth centuries, directly attests to the incorporation of local people into the populace of the colony, as this person is best modeled with qpAdm as deriving his ancestry from the contemporaneous Sicilian_IA group (P = 0.695). Potential social stratification has been suggested associated with the burial sites, as individuals interred in the East necropolis show lower δ13C and δ15N values and higher prevalence of skeletal pathology than those at the West necropolis, indicating a different access to resources (62). However, due to the limited number of samples, we cannot test whether such a significant stratification existed with respect to genetic ancestry.
|
|
