If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The Y-chromosomal haplotype and haplogroup distribution of modern Switzerland still reflects the alpine divide as a geographical barrier for human migration
This is the first detailed study on human Y-chromosomes from Switzerland.
•
606 Swiss male individuals were typed for 27 Y-STRs and 34 Y-SNPs.
•
Significant intra-national differences in haplogroup distribution were revealed.
•
DYS533 allele 12.1 appears to be associated with haplogroup I2 (xM223, xP37.2).
•
Concordance between YFilerPlus® and PowerPlex® Y23 was checked.
Abstract
A sample of 606 Swiss individuals has been characterized for 27 Y-STR and 34 Y-SNPs, defining major European haplogroups. For the first time, a subsample from the southernmost part of Switzerland, the Italian speaking canton Ticino, has been included. The data reveals significant intra-national differences in the distribution of haplogroups R1b-U106, R1b-U152, I1 and J2a north and south of the alpine divide, with R1b-U152 being the most frequent haplogroup among all Swiss subpopulations, reaching 26 % in average and 53 % in the Ticino sample. In addition, a high percentage of haplogroup E1b1b-M35 in Eastern Switzerland corresponds well with data reported from Western Austria. In general, we detected a low level of differentiation between the subgroups north of the alpine divide. The dataset also revealed a variety of microvariants. Some of them were previously known to be associated with particular haplogroups. However, we discovered one microvariant in DYS533 that seems to be closely associated with haplogroup I2-P215 (xM223). This association had not yet been reported to date. The concordance study with two STR-kits suggests that the DYS533 microvariant is due to an InDel in the flanking regions of the marker. One individual carried a large deletion, frequently detected in people of East Asian ancestry, encompassing the amelogenin locus. To our knowledge, this is the first time that such a deletion has been observed within European haplogroup R1b-U152. This is the first comprehensive Y chromosomal dataset for Switzerland, demonstrating significant population substructure due to an intra-national geographical barrier.
The utility of Y chromosome population genetics is twofold: First of all, Y-STR profiles are of tremendous value for forensic investigations. They are frequently used to separately interpret male components in biological traces recovered from victims of sexual abuse, where classical autosomal DNA profiles are not promising, because of excess female cell material. In addition, Y-chromosomal data from a crime scene can also be used to trace down a possible perpetrator by family links [
]. For proper interpretation of DNA from crime scenes, numerous population genetic studies and a large and reliable database are necessary. Such a database exists with the Y Chromosome Haplotype Reference Database (YHRD) [
], however needing continuous contributions. Second, because of the lack of recombination, the genetic data on the Y chromosome can be used to trace back historical events of human migration. The Y chromosome as lineage marker has contributed a good share to the revelation of the mechanisms underlying the geographical structuring of present day human genetic diversity [
With the present paper we would like to contribute to both fields of Y chromosome application. We characterized 606 Swiss male individuals for 27 Y-STRs, frequently used in forensic genetics, using two different multiplex PCR kits, YFiler Plus® and PowerPlex Y23®. Forensically relevant parameters were measured and the data was compared to data from neighboring countries. Furthermore, SNaPshot® assays, comprising 34 SNPs in total, were conducted for all samples to determine major European Y haplogroups. We compared the haplogroup distributions from different regional subpopulations among each other and revealed the associations of Y-STR microvariants with their respective haplogroup.
Switzerland has for longtime been a missing piece in the European Y chromosome puzzle. With the present study we present the first comprehensive and systematically collected Y chromosomal population data for the three principal linguistic groups in Switzerland.
2. Materials and methods
2.1 Sample collection
Samples from 606 male Swiss individuals were characterized for the present study. 435 individuals were German speaking, 118 French speaking and 51 Italian speaking. One person reported both, Italian and German as mother tongues, one person’s mother tongue was Romansh. Sizes of the defined regional subgroups were the following: WS (Western Switzerland) = 129, BE (canton of Bern) = 130, NW (Northwestern Switzerland) = 108, CS (Central Switzerland) = 79, TI (canton of Ticino) = 51, SG (canton of St. Gallen) = 109. For further details concerning the sampling procedure, check [
For every sample, 1 μL of whole blood was extracted using the AutoMateExpress™ device with the PrepFiler Express™ kit (both ThermoFisher, USA). DNA extractions were not quantified by default. 0.5 μL of DNA extract was used in multiplex PCR. Quantification was only done for samples not leading to an interpretable profile and DNA input for multiplex PCR was adapted accordingly. We used a 7500 RT PCR System with the Quantifiler® HP Kit (both ThermoFisher, USA) for qPCR. Multiplex PCR was performed in a reduced reaction volume of 12.5 μL for both kits used in this study, PowerPlex® Y23 (Promega, USA) and YFiler™ Plus (ThermoFisher, USA). Capillary electrophoresis was run with a 3500xl genetic analyzer (ThermoFisher, USA) and data interpretation was carried out with Genemapper® ID-X, v1.4 (Thermo Fisher, US). All peaks above 300rfu were considered as true alleles.
2.3 Haplogroup prediction
Haplogroups were predicted to facilitate the selection of primer panels for subsequent SNaPshot® analysis. Next to Whit Atheys haplogroup predictor (HAPEST) [
] we used the information on Y-STR haplotype distributions within haplogroups provided by the YHRD under the feature “Ancestry information”. This program is not a prediction tool but assigns the 24,316 SNP-typed haplotypes within the YHRD [
] (as of Release 61, June 24, 2019) to branches of the phylogenetic tree (haplogroups). This allows a prediction of the most probable haplogroup assignment of haplotypes by comparison of the new, yet unclassified STR haplotypes with the registered SNP-typed haplotypes from the YHRD. A reduction of loci in the panel increases the haplotype count in the database and the chance to receive a haplogroup frequency. The reduction of the haplotype however, may increase the chance to find haplotypes which are not “identical by descent” (IBD) but “identical by state” (IBS) and may therefore apparently fall in different branches of the phylogenetic tree. We followed a hierarchical approach by checking YHRD ancestry information in the following order, based on: YFiler haplotypes, minimal haplotypes, YFiler one-step neighbors and minimal one-step neighbors. To assess the overall prediction accuracy of the two approaches, the most probable predicted haplogroup was compared to the haplogroup revealed by SNP analysis. All predicted haplogroups not in conflict with a more specific haplogroup determined by SNP analysis, were counted as correct. If two different haplogroups were predicted equally probable and one of them was confirmed as correct, the prediction was counted as neither wrong nor right. If the clade predicted is less specific (e.g. J) than the clade determined (e.g. J1), this was counted as neither wrong nor right. According to the SNP typing, the following clades were defined to address the prediction accuracy for this Western European sample: D, E, G, I1, I2, J1, J2, R1b, R1a, R2, O, N, KLT (no Q, R, N, O), F (no G, I, J, KLT).
2.4 SNP typing
33 different haplogroup-defining SNPs and one recurrent SNP [
], always accompanied by a blank negative control and 2800 M DNA as a positive control. Samples were run on an Applied Biosystems® 3130xl Genetic Analyzer with 36-cm capillaries and POP 4 polymer (Thermo Fisher, USA). Data interpretation was carried out with Genemapper® ID-X, v1.4 (Thermo Fisher, USA). Primers used in this study are listed in Supplementary Table 1. Supplementary Fig. 1 shows a schematic tree of the Y-SNPs analyzed in the present study. SNP information was gathered from YHRD [
Gemeinsame Empfehlungen der Projektgruppe „Biostatistische DNA-Berechnungen “und der Spurenkommission zur biostatistischen Bewertung von Y‑chromosomalen DNA-Befunden.
]. For 11 samples, no DL value could be estimated, due to duplicated, intermediate or null alleles in the YFiler marker subset. Interregional FST values were calculated using STRAF [
], excluding the markers DYS385a/b and DYF3871S1. For analysis with STRAF, DYS389I has been subtracted from DYS389II and intermediate or duplicated alleles were set as null. Principal component analysis of haplogroup I2 was also done with STRAF. The 12.1 allele in DYS533 was not used. Instead we used allele 12, as detected with PowerPlex® Y23. The multiple dimensional scaling (MDS) plot for national datasets was generated by the respective YHRD tool [
], based on RST values for the PowerPlex® Y23 panel and 10,000 permutations. Information about the composition of the different national datasets can be retrieved from https://yhrd.org/pages/resources/national_databases. Sizes of PPY23 datasets from YHRD release 61 were the following: Austria (259 haplotypes), Belgium (728), Denmark (185), Germany (2821), Hungary (591), Ireland (823), Italy (1939), Spain (822). 17 haplotypes were removed for AMOVA and MDS, due to duplicated, triplicated, intermediate or null alleles. Fisher’s exact test for significance of the differences in haplogroup distribution was calculated using R [
], using a Gaussian radius multiplier of 0.4 and 13 shades. The dataset was submitted to YHRD for quality control, as required by the ISFG guidelines [
] and given the six accession numbers YA004658 to YA004663. The haplotypes are included in YHRD release R62. For all haplotypes with discordances between the two kits (e.g. 12.1 and 12 in DYS533), the full allele and not the intermediate allele has been registered on YHRD. The full dataset can be requested from the corresponding author of this paper, for academic research purpose only.
3. Results
3.1 STR haplotype diversity and DL frequency estimation results
The haplotype diversity of the 606 Swiss male individuals for different marker subsets is shown in Table 1. Haplotype diversity was identical for the YFilerPlus® subset and the maximal marker set, additionally including the two markers DYS549 and DYS643 from PowerPlex Y23®. The distribution of haplotype frequencies, estimated with the Discrete Laplace method [
] in the YHRD metapopulation "Western European" based on YFiler haplotypes are depicted in the boxplot in Fig. 1a. The lowest frequency for this metapopulation has been capped to 1 in 73,136,090 haplotypes [
]. The most common haplotype was estimated to 1 in 198. The median frequency is at 1 in 563,526. Fig. 1b illustrates the distribution of the ratios of the estimated haplotype frequencies in the global YHRD population and in the YHRD metapopulation "Western European". For explanation: a ratio of 6 means that the respective haplotype is estimated 6 times more frequent in the "Western European" metapopulation than in the global population. For 90 % of the haplotypes, this ratio is ≥1. For 95 % of the haplotypes the ratio is ≥0.5 and only 7 haplotypes (1.2 %) have a ratio less than 0.1, meaning that they are estimated more than 10 times less frequent in the "Western European" metapopulation than in the global population.
Table 1Haplotype diversity for different marker sets.
Fig. 1A) Haplotype frequency distribution as estimated with DL for the YFiler marker set, displayed as reciprocal values. The median frequency is at 1 in 563‘526. B) Distribution of the ratios (pWE / pG) between the estimated haplotype frequency in the “Western European” metapopulation (pWE) and the estimated haplotype frequency in the global population (pG).
Pairwise FST calculations based on the maximal STR marker set (YFilerPlus® + PowerPlex Y23®) show little intra-national differentiation among the 6 regional subpopulations (Table 2). In line with our previous observations [
], all subpopulations show the largest FST values in pairwise comparison with the southernmost Swiss canton Ticino subpopulation, with the largest difference being the one between Northwestern Switzerland and Ticino. We also compared our dataset to datasets from other countries, using the AMOVA tool from YHRD [
]. The multiple dimensional scaling plot in Fig. 2a localizes the Swiss data between the datasets from neighboring countries. For one of the direct neighboring countries, no data was included, since there was no French dataset for PowerPlex® Y23 available on the YHRD. If we divide the sample into language subgroups, the German speaking subpopulation locates even closer to the Austrian sample, whereas the French speaking subpopulation is somewhat closer to the samples from Belgium and Spain. Surprisingly, the Italian speaking sample co-localizes with the sample from Spain and is significantly different from the Italian sample, registered on YHRD. As a control, we also checked the genetic distance of our regional subsamples to the four other Swiss YFiler datasets registered on YHRD. They show all a high degree of similarity, except for the sample from Basel that exhibits extremely large RST values towards all the other subpopulations, ranging from 0.084 to 0.173, even towards the sample from the same region of Northwestern Switzerland (0.101). All RST values and corresponding p-values generated with the YHRD AMOVA tool are available in Supplementary Table 2.
Table 2Pairwise FST values among the different regional subgroups. NW = Northwestern Switzerland; CS = Central Switzerland; BE = Bern area; TI = Ticino; WS = Western Switzerland; SG = St. Gallen.
Fig. 2Multiple dimensional scaling blot based on RST values, generated for PowerPlex® Y23 datasets with the AMOVA tool from YHRD. A) Comparison of the whole sample ("Switzerland") to other national European datasets, registered on YHRD. B) Comparison of Swiss ("CH") subpopulations based on mother tongue to national datasets from neighboring countries. Spain was included as the next western country in mainland Europe, since no French sample for PowerPlex® Y23 was available on the YHRD. The data points for Spain and the Italian speaking Swiss subsample collapse into one.
3.2 Kit concordance, dropouts, multi-allelic patterns and intermediate alleles
All discordances between the two multiplex kits, deletions, copy number variants and intermediate alleles are listed in Table 3. The table also lists the haplogroup of the respective sample(s). Out of the 606 samples, 32 showed at least one particularity, leading to an event rate of 5.3 %. The two multiplex kits PowerPlex® Y23 and YFilerPlus® share 20 markers. Discordant allele values were detected for 8 samples, leading to a rate of discordant profiles of 1.3 %.
Table 3Discordances, dropouts, multi-allelic patterns and intermediate alleles. Two samples, marked with */** are concerned by two events. R1b-M269 is equivalent to R1b-M269 (xU152, xU106), I2-P215 is equivalent to I2-P215 (xM223, xP37.2), with P37.2 being a recurrent mutation.
The most frequent discordance is an intermediate allele 12.1 in DYS533, only detected with the YFilerPlus® kit. In the present dataset, this intermediate allele occurs exclusively in an I2-P215 background. In addition, all five samples show the ancestral version of M223 and the ancestral version of the recurrent SNP P37.2 [
]. All other individuals (n = 35) typed as I2-P215 show either a derived version of P37.2 or of M223. Principal component analysis of all samples belonging to haplogroup I2 shows a distinct cluster of those 5 samples, when the third component is included (Fig. 3).
Fig. 3Principle component analysis of the 40 samples belonging to haplogroup I2, showing all correlations between the three first components (see titles on axes). The five samples with the 12.1 allele in DYS533, positive for I2-P215 but ancestral for M223 and the recurrent SNP P37.2, form a distinct cluster (orange), separate from the samples with derived SNPs for M223 (blue) and P37.2 (brown) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
One of the samples carries a larger chromosomal deletion, as shown by the simultaneous dropout of the six neighboring loci DYS576, DYS627, DYS458, DYS570, DYS449, DYS481 and Amelogenin, all located on the short arm of the Y chromosome.
3.3 Success rates of haplogroup predictions
To specifically target our SNaPshot analysis for Y haplogroups, we first predicted the haplogroups by two approaches: a search for the same or similar haplotypes, already typed for SNPs, on the YHRD [
]. Subsequent comparison with the SNaPshot data enables us to make a statement on the accuracies of the predictions. Overall the haplogroup predictions were quite reliable, as can be seen from Table 4. Whereas Whit Athey’s gives a prediction for every haplotype entered, the “prediction rate” based on YHRD data depends on whether any similar haplotype has already been registered and SNP typed or not. It has to be mentioned that all predictions concordant between YHRD and Whit Athey’s proofed to be right. Also, no wrong haplotypes were assigned when ancestry information could be retrieved from YHRD for the YFiler dataset. However, ancestry information from the YFiler dataset on YHRD could be retrieved for less than 50 % of the searched samples, due to a lack of registered SNP data.
Table 4Haplogroup prediction rates and prediction accuracies for Whit Athey’s haplogroup predictor and the different YHRD ancestry information queries. Neighbors are 1-step neighbors, differing from the haplotype used for the query in one repeat, in one marker. The prediction rate is the percentage of all samples for which a prediction (meaning a retrieval of ancestry information) was possible. This rate is restricted by the available SNP data for comparison on the YHRD.
] were maintained. All haplogroup proportions are listed in Table 5. We detected five significant regional differences in haplogroup spread. Whereas the haplogroups I1-M253 and R1b-U106 are more or less evenly distributed north of the Alps, they are almost absent from the Ticino sample. In return, haplogroups J2a-M410 and R1b-U152 are far more abundant in the Ticino sample than in the rest of the country. Furthermore, we detected a significant enrichment of haplogroup E1b1b-M35 in the easternmost sample from St. Gallen. We could also detect a slightly larger proportion of E1b1b-M35 in the sample from Western Switzerland. However, this observation proofed not to be significant.
Table 5Y Haplogroup distribution among regional subgroups (values given in %). Columns not summing up to 100 % are due to rounding errors. Significant differences in the Ticino and St. Gallen sample compared to the distribution over the rest of Switzerland are *p < 0.05, **p < 0.01 and ***p < 0.0001 (Fisher’s exact test). Abbreviations and absolute number of samples from the respective region: NW = Northwestern Switzerland (n=108); CS = Central Switzerland (n=79); WS = Western Switzerland (n=129); SG = St. Gallen (Eastern Switzerland; n=109); BE = Bern area (n=130); TI = Ticino (n=51); Hg (tot) = percentage of the respective haplogroup in the total Swiss sample of 606 individuals.
As expected, we observe a good correspondence of the dataset with the metapopulation “Western European”, what can be concluded from the distribution of the estimated haplotype frequencies. The fact that 90 % of the haplotypes are predicted to be more frequent in the Western European than in the global panel, can be seen as a successful quality control of the sampling scheme. The population sample also fits well in the context of the neighboring countries and shows no noteworthy differences compared to the Swiss datasets previously registered on the YHRD. The only exception concerns the sample from Basel. However, since the Basel sample on YHRD shows large genetic differences with all other Swiss samples, including our sample collected from the same region, we assume some kind of sampling error for this regional subsample and we would like to suggest that it should be used with caution for any interpretation and comparison. The fact that the Italian speaking subsample co-locates rather with the sample from Spain than with the sample from Italy (Fig. 2), might be attributed to the higher overall percentage of haplogroup R1b in Spain than in Italy [
]. The fraction of R1b in the Spanish population corresponds better to the 70 % R1b in the Ticino sample.
4.2 Microvariants and concordance
We observed a variety of intermediate alleles among the 606 individuals (Table 3). Some of the detected variant alleles have previously been described as being associated with particular haplogroups. The .2 alleles in DYS458 are usually associated with haplogroup J1 [
], as it is the case for our Swiss population sample. The kit-dependent variant allele 25.1 in DYS481, being caused by a SNP in the flanking region of the marker, has been shown to be associated with haplogroup R2 [
Y-chromosome short tandem repeat intermediate variant alleles DYS392.2, DYS449.2, and DYS385.2 delineate new phylogenetic substructure in human Y-chromosome haplogroup tree.
]. The same tri-allelic pattern for DYS385a/b as we could observe here has been previously reported in a sample from Eastern Africa, also within a haplogroup E background [
We observed one kit dependent dropout in DYS390 and a null allele for DYS448 in the same individual. Whereas we could not find a reported case for the DYS390 dropout, deletion or dropout due to primer binding site mutations of DYS448 has been described previously [
]. We assume that the signal dropout for DYS448 is rather due to a deletion than to a primer binding site mutation, since we observe strictly no CE signal with both multiplex kits. The locus DYS448 is located in an area of the Y-chromosome prone to deletions and recombination due to repeat structures [
Dynamic nature of the proximal AZFc region of the human Y chromosome: multiple independent deletion and duplication events revealed by microsatellite analysis.
Dynamic nature of the proximal AZFc region of the human Y chromosome: multiple independent deletion and duplication events revealed by microsatellite analysis.
]. In our study, this large deletion surprisingly occurred within an R1b-U152 haplogroup background, a haplogroup that is thought to have a relatively recent origin in Europe [
]. To our knowledge, this is the first time that such a deletion is reported for an individual of European paternal ancestry.
Concordance of multiplex kits is crucial, whenever profiles are generated by different labs for comparison. PowerPlex® Y23 and YFilerPlus® are two Y-STR kits that are probably the most widely used in the field. In our study, the two kits exhibit a good concordance rate. Of the 8 samples, showing different allele values between the two kits, 5 displayed the same .1 allele with YFilerPlus® in DYS533. This allele could be due to an InDel in the flanking regions of the marker. Such an InDel has been previously described by Huszar et al. [
], yet in another Y haplogroup background (C1a). All five samples from our study occurred in an I2-P215 (xM223, xP37.2) background, what has not yet been reported to date. The discordant DYS533 allele 12.1 might therefore carry valuable phylogenetic information. However, the allele occurred at a relatively high frequency in our sample and might be problematic for Y-STR databases, since it remains undetected with the widely used kit PowerPlex® Y23. A quick search on the YHRD returned a total of 24 intermediate .1 alleles for DYS533, spanning the allelic range from 11.1 to 14.1.
4.3 Haplogroup prediction
Given the dubious reputation of the prediction tools [
], we were surprised how well the haplogroup predictions corresponded to the haplogroups determined by SNaPshot assay. So, even though we would agree that for reliable results, every SNP should be finally determined in the wet lab, we cannot deny that for samples of Western European ancestry, predictors seem to deliver good preliminary results. The HAPEST predictor we used here has already been shown to deliver accurate predictions for typical European haplogroups [
]. We assume that such a high accuracy of haplogroup prediction of 95 % could be achieved only because we have very good data coverage for Western Europe. For most reliable predictions, we recommend combining an YHRD search with the haplogroup prediction tool. All haplogroups that were concordant between YHRD ancestry information and HAPEST haplogroup prediction turned out to be correct.
4.4 Haplogroup distributions
SNP typing for common European haplogroups revealed some expected patterns, demonstrating that the modern Swiss population still reflects the Alps as geographical barrier for human migration. We detected significantly less haplogroup I1-M253 south of the alpine divide than in the German and French speaking parts of Switzerland (Table 5, Fig. 4a). This was expected, since I1 is most common in Northern Europe and can only be found in small proportions south of the Alps. It has been suggested, that haplogroup I1 has been spread in Scandinavia by migration from Western Europe after the last glacial maximum [
Also not surprising is the significant enrichment of haplogroup J2a-M410 in the Ticino sample (Table 5, Fig. 4b). J2 probably originates from the Near East [
Origin, diffusion, and differentiation of Y-chromosome haplogroups E and J: inferences on the neolithization of Europe and later migratory events in the Mediterranean area.
We also detected significant differences for the Ticino sample compared to the rest of Switzerland for two sub-lineages of R1b-M267. R1b-M267 is the most frequent haplogroup in Western Europe, with increasing proportions from east to west [
]. We detected significant differences in the distribution of two sublineages of R1b-M269 north and south of the Alps: notably lineages R1b-U106 and R1b-U152 (Fig. 4c and d). R1b-U106 is mainly spread along the river Rhine, reaching the largest proportions at the southern coast of the North Sea [
]. It has been suggested that it originates from a Franco-Cantabrian region and has been brought to the Alps and northern Italy by migration along the Mediterranean coast [
]. In this pre-alpine region, located about 50 km from Ticino, the percentage of individuals with haplogroup R1b-U152 is around 50 %, just as for our Ticino sample (Table 5). This local present day hotspot for haplogroup R1b-U152 fits quite well the ancient habitats of Celtic cultures such as the Lepontii, being most probably at the origin of the name “Valle Leventina” for the upper part of the Ticino Valley, or the Orobii [
The Linguistic Position of Lepontic, Proceedings of the Twenty-Fourth Annual Meeting of the Berkeley Linguistics Society: Special Session on Indo-European Subgrouping and Internal Relation, Linguistic Society of America1998: 2-11
]. Haplogroup R1b-U152 is significantly less frequent north of the Alps, but remains the most frequent haplogroup throughout the entire country (Table 5). Jörg Jenatsch, a Swiss national hero from the 17th century, also belonged to haplogroup R1b-U152 [
We could also observe an uneven distribution of haplogroup E1b1b-M35 (Fig. 4e). However, this one does not follow a gradient from north to south but manifests as an enrichment of haplogroup E1b1b-M35 in the eastern part of the country, close to Austria. This pattern fits quite well with data from Tyrol, a region in Western Austria [
Differences in urbanization degree and consequences on the diversity of conventional vs. rapidly mutating Y-STRs in five municipalities from a small region of the Tyrolean Alps in Austria.
], showing an important fraction of 16.9 % haplogroup E in this area, mostly attributed to E-M78, a subclade of E1b1b-M35.
5. Concluding remarks
In sum, we are presenting the first detailed study of the Y-chromosomal landscape in Switzerland. The outcome fits very well the expectations and the Y chromosomal data from neighboring regions. The modern Swiss population still mirrors the alpine divide as an important barrier for human migration on an intra-national level.
Ethics statement
All samples were collected with informed written consent. They were reversibly anonymized, to permit the donors to exert their right to withdraw their sample at any time. The Institute of Forensic Medicine, University of Bern, obtained the samples under an arbitrary number. The written consent documents with the names of the donors remained with the Red Cross. All documents distributed to the donors upon sampling were submitted to the responsible cantonal ethical committee and approval obtained.
CRediT authorship contribution statement
Martin Zieger: Writing - original draft, Writing - review & editing. Silvia Utz: Resources.
Acknowledgements
We thank all the donors for participating in the project and Ina Krebber (Interregionale Blutspende SRK), for her help in organizing the sample collection. Special thanks go also to Ludmilla Lieder for technical assistance, as well as to Maria Geppert and Jessica Rothe (both Institute of Legal Medicine and Forensic Sciences, Charité - Universitätsmedizin Berlin) for help with the SNaPshot multiplex assays. We would also like to thank Sandra Lösch (Anthropology Dept., Institute of Forensic Medicine, University of Bern) for helping out with literature about Swiss Iron Age populations. Finally M.Z. would especially like to thank Lutz Roewer (Institute of Legal Medicine and Forensic Sciences, Charité - Universitätsmedizin Berlin) for hosting him in his lab and for discussing the present manuscript.
Appendix A. Supplementary data
The following are Supplementary data to this article:
Gemeinsame Empfehlungen der Projektgruppe „Biostatistische DNA-Berechnungen “und der Spurenkommission zur biostatistischen Bewertung von Y‑chromosomalen DNA-Befunden.
Y-chromosome short tandem repeat intermediate variant alleles DYS392.2, DYS449.2, and DYS385.2 delineate new phylogenetic substructure in human Y-chromosome haplogroup tree.
Dynamic nature of the proximal AZFc region of the human Y chromosome: multiple independent deletion and duplication events revealed by microsatellite analysis.
Origin, diffusion, and differentiation of Y-chromosome haplogroups E and J: inferences on the neolithization of Europe and later migratory events in the Mediterranean area.
The Linguistic Position of Lepontic, Proceedings of the Twenty-Fourth Annual Meeting of the Berkeley Linguistics Society: Special Session on Indo-European Subgrouping and Internal Relation, Linguistic Society of America1998: 2-11
Differences in urbanization degree and consequences on the diversity of conventional vs. rapidly mutating Y-STRs in five municipalities from a small region of the Tyrolean Alps in Austria.
30118-6&id=gr1.jpg){kind=link}
30118-6&id=gr2.jpg){kind=link}
30118-6&id=gr3.jpg){kind=link}
30118-6&id=gr4.jpg){kind=link}