|
Post by Admin on Jul 14, 2023 19:43:35 GMT
2. Materials and methods During rescue excavations done in 2007, a burial site was unearthed in Ora/Auer by the Archaeological Office of the Autonomous Province of Bolzano/Bozen. As shown in Fig. 2a, Ind. A was originally found lying on the left side in a crouched position. The taphonomic analysis suggested that this inhumed individual underwent various phases of displacement inside the grave due to soft tissue decomposition, gravity effects, and small-sized animals (rodents) that caused disarticulation of the joints (Rizzi et al. 2010). Ind. B (Fig. 2a) was buried in contracted position, but differently from the other individual, he was lying on the right side. Ind. B moreover presented clear signs of rodent tooth-marks on the surface of some bones. The immature skeletal remains of Ind. C were most affected by taphonomic agents, particularly by rodents “which had an erosive action” on the bones (Rizzi et al., 2010: 41).
For the paleogenomic analysis, left petrous parts (PP) of the temporal bones still anatomically connected to the cranium (Fig. 2b) were sampled from both adult individuals.
For each sample, approximately 150 mg of powdered bone was collected from the inner part of the PP (Fig. 2b) using a drill (Pinhasi et al., 2015) in a dedicated pre-PCR area of the aDNA laboratory of Eurac Research in Bolzano (Italy) following all stringent rules required for aDNA analyses. DNA samples were then extracted using a purification method based on silica columns (Gamba et al., 2014) and double-stranded genomic libraries were constructed (Meyer and Kircher, 2010). These were then sent to an external company (Macrogen) for shotgun sequencing [100 bp paired-end (PE), Hiseq2500 and 150 bp HiSeq-X systems, Illumina]. This first molecular screening was finalized to assess the authentication and good preservation of aDNA from the samples. These were then enriched for more than 2 million polymorphic sites in the human genome using the in-solution target capture kit myBaits ® Expert Human Affinities – Prime Plus (Arbor Bioscience). This includes three different target sets: 1) Prime 1240 K includes 1.24 million population-informative SNPs (Single Nucleotide Polymorphisms); 2) Y Chromosome 46 K targets sites on the Y Chromosome; and 3) MitoTrio probes which cover the complete mtDNA genome. For both samples, the amount of library input for the enrichment reached 1000 ng – the quantity recommended in the protocol for successful enrichment (Human Affinities, Version 1.0, 2021 – Daicel Arbor Bioscience). This procedure was adjusted in terms of hybridization period by increasing the time to 40 h instead of 16 h suggested by the protocol.
Bioinformatic analysis of the sequenced reads (shotgun and capture data) was performed. Reads were trimmed and merged (PEAR) (Zhang et al., 2014) if they overlapped by at least 25 bp and had a minimum length of 25 assembled sequences. The QualityFilterFastQ.py script (Kircher, 2012) was applied to eliminate reads with 5 bases below the quality threshold of 15. Reads were then aligned to the Genome Reference Consortium Human Build 37 (hg19) and revised Cambridge Reference Sequence (rCRS) (Andrews et al., 1999) with BWA (Li and Durbin, 2010) using a minimum mapping quality set at 25. Duplicates were removed by using Dedup (Peltzer et al., 2016). Damage patterns among the ancient reads were tracked and quantified (fragmentation and misincorporation patterns) by using mapDamage (Jónsson et al., 2013). Contamination estimates based on mtDNA data were inferred by using Schmutzi (Renaud et al., 2015). Those on X-chromosome data were estimated applying the method implemented in ANGSD (Analysis of next generation Sequencing Data) (Rasmussen et al., 2011).
The two bam files obtained for each sample from molecular screening and enrichment were then merged for a total of 552,688 (Ind. A) and 1.019,831 (Ind. B) SNPs sites hitting on the 1.240 K panel and were used for kinship analyses.
This was performed using three different methods: READ, TKGWV2 and lcMLkin. The method implemented in READ (Relationship Estimation from Ancient DNA; Kuhn et al., 2018) calculates and normalizes pairwise mismatch rates in non-overlapping windows across the whole genome (pseudo-haploid data). The normalization was carried out using standard parameters (i.e. median of all average P0s). The other two methods, TKGWV2 (Thomas Kent Genome-Wide Variants 2; Fernandes et al., 2021) and lcMLkin (Lipatov et al., 2015), use genotype likelihoods and population allele frequencies to infer genetic relatedness between individuals. In both cases, analyses were performed with default parameters. TKGWV2 is used to infer kinship up to the 2nd degree and lcMLkin to the 5th degree.
Molecular sex determination was inferred by calculating the ratio of the total merged sequences aligning to the X and Y Chromosomes (Skoglund et al., 2013).
Y Chromosomal haplogroups were assigned using Yleaf software v2.2 (Ralf et al., 2018) with standard settings, while classification of the mtDNA haplogroups was performed by applying Haplogrep 2.4.0 (Kloss-Brandstätter et al., 2011) and the most recent up to date phylogenetic tree of worldwide human mitochondrial DNA variation.
The recovery, sampling and all the analyses performed in this study were authorized by the competent authority (13.2 Archaeological Office, Autonomous Province of Bolzano/Bozen, Italy).
|
|
|
Post by Admin on Jul 16, 2023 17:22:42 GMT
3. Results and discussion Molecular screening revealed good preservation and quality of aDNA for both adult individuals. Bioinformatic analyses of shotgun sequencing reads showed typical damage patterns for aDNA and low contamination from modern human DNA using mtDNA data (average 3%). The percentage of reads mapped to the human reference genome (hg19) was 7.3% and 24.8% for Ind. A and Ind. B, respectively (Table SM1 and Table SM2). Additionally, bioinformatic analysis of sequencing reads obtained after enrichment showed very good human endogenous content in both samples (>92%) and mean coverage on 1240 K sites of 0.448 in Ind. A and 0.827 in Ind. B. Low contamination from modern mtDNA was confirmed (average 1–2%) and is based on a higher coverage of the mitogenomes (31.5X and 172X for Ind. A and Ind. B, respectively). Low contamination estimates (Ind. A, 1.9–2.4%; Ind. B, 1–1.1%) were also inferred using nuclear X chromosome data (Table SM3). The final mean genome coverage based on merged data was 0.07 X and 0.34 X for Ind. A and Ind. B, respectively (Table SM4).
Successful recovery of aDNA from both individuals made it possible to answer our two main questions.
First, ancient molecular analysis confirmed that the two adults were biologically male (XY). Molecular sex estimations were based on several human reads that were much higher (1,262,31 for Ind. A and 4,250,379 for Ind. B) than the number needed by the method to obtain a reliable estimate (about 100,000 human sequences; Skoglund et al. 2013; Table SM2). Therefore, any previous speculation of their sex estimations made by Rizzi and her colleagues was clarified. This highlights that presence of an infant in a multiple burial does not necessarily imply the presence of a female.
There are other examples cited in the literature in which the biological sex of ancient individuals diverged from the initial assumption based on archaeological records. This is the case for individuals buried in double or multiple tombs, particularly when the bodies are found placed next to each other. In these cases, the natural hypothesis is to presume that they are a man and a woman. For instance, the two “Lovers of Mantova” (Italy), which were intentionally buried hand-in-hand, were therefore considered a loving couple (Vazzana et al., 2018). However, an Amelogenin protein analyses of tooth samples from these two individuals revealed that they were both males (Lugli et al., 2019). Another example is seen where several couples were buried in a Bronze Age double grave in Spain. In this case, they were believed to be married couples, corroborating the existence of heterosexual monogamous marriages during that time period (Castro et al., 1993). However, recent radiocarbon dating has shown that the individuals belonged to different generations, demonstrating the deceased were more likely descendants of one another and not married couples (Esparza et al., 2017). Our study, together with these cases, illustrate the importance of critically reconsidering gender perspectives that have been applied to past societies, as was recently pointed out by Turek (2019).
The second aim of this work was to investigate whether the two individuals from Ora/Auer were genetically related. The application of three methods to infer relatedness based on genomic data identified a first-degree relationship between the two males (e.g., father-son, brothers). Additionally, one method (lcMLkin) specifically detected a parent-offspring relationship (Table 1 and Table SM5). Moreover, data from the unilinear transmitted markers, which are almost exclusively transmitted by only one parent to their offspring, indicated kinship at paternal level. Indeed, Y-Chromosome haplogroup analysis found the same lineage in both males (haplogroup G2a2b2a1a1b1) while those based on the mtDNA data unambiguously assigned two different maternal haplogroups for Ind. A and Ind. B (K1a and J1c + 16261 + 189, respectively). This moreover suggests a different genetic history at maternal level for the two individuals, the further investigation of which is beyond the scope of this paper (Table 1 and Table SM4). Taking all the genetic results into consideration, we exclude the possibility that the two men were brothers, thus most likely indicating that they were father and son.
Table 1. Summary of the genetic results for the two adults found in Ora/Auer. Abbreviations: Anthro ID = Anthropological ID; 1240 K Sites (All) = number of total sites hitting on the 1240 K dataset. Mol Sex = Molecular sex assignment; Haplogroups of the mitochondrial DNA (mtDNA) and Y-Chromosome. Kinship results obtained by applying three methods (lcMLkin, READ and TKGWV2) and final kinship interpretation based on all data.
Anthro ID 1240 K Sites (All) Mol Sex mtDNA haplogroup Y-Chromosome haplogroup Kinship (lcMLkin) Kinship (READ) Kinship (TKGWV2) Relation Ind. A 552.688 XY (Male) J1c + 16261 + 189 G2a2b2a1a1b1 Parent-offspring 1st degree Ind. B 1.019,831 XY (Male) K1a G2a2b2a1a1b1 1st degree Father-Son
|
|
|
Post by Admin on Jul 19, 2023 5:27:51 GMT
The age-at-death estimation performed by Rizzi and colleagues (2010) could also be consistent with a father-son relationship. Indeed, there was only a ten-year age difference between the two men (Ind. A ∼ 30 years old, Ind. B 40+ years old). However, considering that the anthropological age-at-death is the result of an estimation, the gap could also have been greater, up to 20 years or more.
Radiocarbon dating ranges also support the almost contemporaneity of the two relatives.
4. Conclusions The effectiveness of interdisciplinary dialogue combining archaeology, anthropology, and paleogenomics is well represented in this study. We successfully obtained the genetic sex and biological relatedness of two Copper Age adult individuals, thus answering some questions remaining from previous anthropological and taphonomic investigations. This study also shows the important role that paleogenomics plays in better understanding prehistoric funerary contexts which would otherwise be difficult to interpret based on archaeological data alone. This study is part of an ongoing palaeogenomic project performed on several ancient human remains from the eastern Italian Alps in order to broaden our understanding on the genetic and social structure of prehistoric and protohistoric individuals from this alpine region.
Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements The Authors would like to thank Sofia Alemanno for helping in the genetic analysis, Kathrin Renner for editing the map in Fig. 1a, Jasmine Rizzi for the photo of the excavation in Fig. 1b, David E. Michalik for the English language revision, and to our colleague Marco Samadelli for having photographed the cranium in Fig. 2b. The computational results of this work were performed using the Life Science Compute Cluster (LiSC) of the University of Vienna.
|
|
|
Post by Admin on Aug 7, 2023 18:12:17 GMT
Ancient genomes reveal structural shifts after the arrival of Steppe-related ancestry in the Italian Peninsula Summary Across Europe, the genetics of the Chalcolithic/Bronze Age transition is increasingly characterized in terms of an influx of Steppe-related ancestry. The effect of this major shift on the genetic structure of populations in the Italian Peninsula remains underexplored. Here, genome-wide shotgun data for 22 individuals from commingled cave and single burials in Northeastern and Central Italy dated between 3200 and 1500 BCE provide the first genomic characterization of Bronze Age individuals (n = 8; 0.001–1.2× coverage) from the central Italian Peninsula, filling a gap in the literature between 1950 and 1500 BCE. Our study confirms a diversity of ancestry components during the Chalcolithic and the arrival of Steppe-related ancestry in the central Italian Peninsula as early as 1600 BCE, with this ancestry component increasing through time. We detect close patrilineal kinship in the burial patterns of Chalcolithic commingled cave burials and a shift away from this in the Bronze Age (2200–900 BCE) along with lowered runs of homozygosity, which may reflect larger changes in population structure. Finally, we find no evidence that the arrival of Steppe-related ancestry in Central Italy directly led to changes in frequency of 115 phenotypes present in the dataset, rather that the post-Roman Imperial period had a stronger influence, particularly on the frequency of variants associated with protection against Hansen’s disease (leprosy). Our study provides a closer look at local dynamics of demography and phenotypic shifts as they occurred as part of a broader phenomenon of widespread admixture during the Chalcolithic/Bronze Age transition. Introduction The Italian Chalcolithic (or Copper Age [CA]; 3600–2200 BCE), the period between the Late Neolithic (N) (7000–3600 BCE) and the Bronze Age (BA) (2200–900 BCE), is characterized by the development of new tools from different metallic sources and was followed by major cultural transformations, including that of burial practice—from an emphasis on the collective to the individual and of personal, prestige grave goods.1,2 Ancient DNA (aDNA) studies have highlighted the occurrence of major shifts in the genetic profiles of populations coinciding with material culture changes, such as from hunting-gathering to farming.3, 4, 5, 6 At the beginning of the transition from the Chalcolithic to the BA ∼5,000 years ago, people from the Eurasian Steppe arrived in Europe, resulting in further admixing with local populations.7, 8, 9, 10 Although these events have been extensively studied in most of Europe4,11,12 and a number of studies on ancient genomes from the Italian Peninsula, Sardinia, and Sicily have been recently published,4,7,9,13, 14, 15, 16, 17 the demographic dynamics of the Chalcolithic/Early BA in the Italian Peninsula are still not well characterized. Though previous studies place the arrival of a Steppe-related ancestry component in Northern Italy9 and in Sicily16 after 2300 BCE, a chronological gap from 1900 to 900 BCE is present, and little is known about the spread of Steppe-related ancestry in Central Italy. In addition, the available data show an Iranian N-related component detected in Sardinia after 900 BCE,16,17 although affinities to Caucasus hunter-gatherers (CHG) and Iran N farmers are present in Central Italian N individuals13 and in Middle BA Sicily,16 at a lower proportion than modern Italians.11 However, although the BA CHG affinity in Sicily is supported by ƒ4 statistics, the evidence for the N CHG influx is less robust. Furthermore, with a few exceptions,18,19 previous surveys have focused primarily on describing ancestral relationships or inferring movement and mixtures of populations at the expense of questions focusing on the social dynamics associated with these events, e.g., evaluating the kinship structure in prehistoric society. aDNA is proving a useful tool for helping to infer past social structures and reproductive behavior (reviewed in Racimo et al.20). In N Europe, several studies have detected a widespread cultural connection of patrilineal social organization21,22 as well as for the BA transition with large-scale, sex-biased migrations,8,23,24 local patterns of patrilocality and female exogamy,18,25 and the influence of cultural diffusion versus migration.9 Although the social implications of these changes are still debated,26 cultural shifts can have an effect on adaptation (e.g., a change in technology leads to a change in diet, leading to selective pressure on metabolism genes). So far, the social structure(s) in Central Italy during the Chalcolithic/BA transition and whether shifts in cultural practices (kinship, patrilocality, and exogamy) correlate with the introduction of Steppe-related ancestry remains unexplored. This may be partially due to the fact that, although there exists a wide variety of burial practices in the Chalcolithic period in Italy, they are often characterized by collective depositions of commingled remains.2 This has made the anthropological analysis of the burial populations and interpretations regarding kinship and social structure difficult; however, high-throughput aDNA sequencing allows for the genetic screening of large numbers of skeletal samples and reconstruction of individuals from disarticulated remains. In addition to reshaping our understanding of the demographic history of the European continent, analyses of ancient genomes from Europe have recently called into question hypotheses regarding the time depth of selection on phenotypic traits in Europe. For example, aDNA has revealed that Mesolithic hunter-gatherers in Europe could have dark skin and blue eyes (a combination rarely seen today)27, 28, 29 and that selection on skin pigmentation occurred in the last 5,000 years.30,31 Other recent work has suggested that selection within genes related to fatty acid metabolism and starch digestion did not take place during the transition to agriculture but instead initiated closer to 2000 BCE following the introduction of Steppe-related ancestry into Europe,31 with the ancestry component itself a possible driver. Another open question, that of the role of pathogens in shaping human genomes, is now starting to be explored using aDNA. One particular pathogen, Hansen’s disease (leprosy), is first seen in paleopathological evidence in the Mediterranean dating to the BA32 and is noted in Central and Northeast Italy by 300 BCE.33 The disease may have been later spread by Roman military movements34 and increased to high numbers in Europe in the Early Medieval Period, but declined by the 15th century CE, and the role of human genetic adaptation in this decline is unknown. There are a number of genetic loci that have been implicated in the manifestation and progression of the infection,35, 36, 37, 38 including one recently discovered using aDNA.39 Here, using an aDNA approach, we investigated the diversity of ancestry components prior to and through the Chalcolithic/BA transition in Northeastern and Central Italy and whether shifts in Steppe-related ancestry correlate with changes in inferred social structure and/or phenotypic traits. www.cell.com/current-biology/fulltext/S0960-9822(21)00535-2
|
|
|
Post by Admin on Aug 17, 2023 17:31:45 GMT
Results We extracted DNA from 51 skeletal elements (teeth = 37, petrous bones = 10, and additional powder from petrous bones = 4) at the Ancient DNA Laboratory of the Institute of Genomics, University of Tartu in Estonia. The human remains are from one necropolis (Necropoli di Gattolino; hereafter, “Gattolino”) and three cave sites located in Northeastern (Grottina dei Covoloni del Broion: “Broion”) and Central Italy (Grotta La Sassa: “La Sassa” and Grotta Regina Margherita: “Regina Margherita”; Figures 1A and S1). After screening 47 libraries at a low depth (±20M reads per library), we identified and sequenced 20 libraries with more than 4% endogenous human DNA and mtDNA-based contamination estimates less than 1.44% (Data S1A and S1B). For the disarticulated remains in the cave sites, we calculated the pairwise mismatch rate of SNPs (P0) on pseudo-haploid data as implemented in READ40 to identify genetically identical samples and attempt to calculate a minimum number of individuals (see STAR Methods and Data S1B for details). Figure 1 Geographical location of samples and relative or absolute dating (A) Map of the geographical location of selected published (smaller, transparent) and newly generated samples from the Italian Peninsula, Sardinia, and Sicily included in this study. The titled locations are here newly reported. See also Table 1 and Data S1. (B) Distribution of relative and absolute dating and genetic assignment from newly generated samples (i) and published (transparent; order: [ii] Italian Peninsula; [iii] Sardinia; and [iv] Sicily). See also Data S1A–S1D The sequences of identical samples were merged together, leaving 22 unique individuals: eleven from Broion (Italy_Broion_CA = 4, Italy_Broion_EBA = 2, and Italy_Broion_BA = 5), four from Gattolino (Italy_Gattolino_CA), three from Regina Margherita (Italy_ReginaMargherita_BA), and four from La Sassa (Italy_LaSassa_CA). The final data are composed of individuals with endogenous DNA between 0.48% and 48.87%, average genomic coverage between 0.0016× and 1.24×, and estimated contamination rates of 0.00%–1.44% (mtDNA-based) and 0.45%–1.98% (X-chromosome-based in males only; Table 1; Data S1A and S1B).
|
|