Temporal and structural genetic variation in reindeer (Rangifer tarandus) associated with the pastoral transition in Northwestern Siberia

Abstract Just as the domestication of livestock is often cited as a key element in the Neolithic transition to settled, the emergence of large‐scaled reindeer husbandry was a fundamental social transformation for the indigenous peoples of Arctic Eurasia. To better understand the history of reindeer domestication, and the genetic processes associated with the pastoral transition in the Eurasian Arctic, we analyzed archaeological and contemporary reindeer samples from Northwestern Siberia. The material represents Rangifer genealogies spanning from 15,000 years ago to the 18th century, as well as modern samples from the wild Taĭmyr population and from domestic herds managed by Nenetses. The wild and the domestic population are the largest populations of their kind in Northern Eurasia, and some Nenetses hold their domestic reindeer beside their wild cousins. Our analyses of 197 modern and 223 ancient mitochondrial DNA sequences revealed two genetic clusters, which are interpreted as representing the gene pools of contemporary domestic and past wild reindeer. Among a total of 137 different mitochondrial haplotypes identified in both the modern and archaeological samples, only 21 were detected in the modern domestic gene pool, while 11 of these were absent from the wild gene pool. The significant temporal genetic shift that we associate with the pastoral transition suggests that the emergence and spread of reindeer pastoralism in Northwestern Siberia originated with the translocation and subsequent selective breeding of a special type of animal from outside the region. The distinct and persistent domestic characteristics of the haplotype structure since the 18th century suggests little genetic exchange since then. The absence of the typical domestic clade in modern nearby wild populations suggests that the contemporary Nenets domestic breed feature an ancestry from outside its present main distribution, possibly from further South.


| INTRODUC TI ON
The domestication of livestock has played a key role in human history. Just as the domestication of sheep, goat, cattle, and pigs were critical elements in the Neolithic transition in many parts of Western Eurasia (Clutton-Brock, 1999;Zeder, 2008), the domestication of reindeer was important for the development and settlement of the Eurasian Arctic (Bjørklund, 2013;Jernsletten & Klokov, 2002;Kofinas, Osherenko, Klein, & Forbes, 2000). In some parts of this region, people held very small groups of highly trained domestic reindeer to facilitate hunting and fishing since the start of the first millennium (Fedorova, 2000;Gusev, Plekhanov, & Fedorova, 2016).
However, what is commonly described as "the pastoral transition" documents a type of large-scale, extensive reindeer husbandry which is thought to represent one of the most fundamental social transformations to ever take place in the Eurasian Arctic region (Hansen & Olsen, 2014;Krupnik, 1993). This pastoral transition established new relations between humans and animals and led to new settlement and land use patterns across large portions of northern Eurasia. Hence, more knowledge of when, where, and how Rangifer pastoralism emerged is key to understand the history of many Arctic communities.
Reindeer (Rangifer tarandus) are distributed throughout most of the northern Holarctic region with the domestic form mainly confined to the Eurasian continent. Among the world's total stock of nearly 2,000,000 domestic reindeer, ~2/3 are distributed in Russia and 1/3 in Fennoscandia (Klokov, 2007). Unlike in the Scandinavian countries, where reindeer husbandry is mainly associated with the indigenous Sámi population, a great variety of indigenous peoples across Russia hold domestic reindeer using a wide variety of reindeer herding strategies. Large-scale nomadic herding, which involves long seasonal migrations and large herds, dominates the Arctic tundra and forest-tundra areas, while in the more southern taiga and mountain areas reindeer herders typically keep small herds of twenty to a few hundred head that are mainly used to facilitate hunting and transport (Klokov, 2007). Beyond the general distinction of tundra and taiga reindeer husbandry forms, domestic reindeer in Russia subdivided into four official breeds: Nenets, Even, Evenk, and Chukot, which are named after the ethnic groups assumed to have established these breeds with their particular traits and adaptations to their respective environment (Zabrodin & Borozdin, 1989).
Despite decades of research across multiple disciplines, many key issues related to the origins, spread, and intensification of reindeer domestication remain poorly understood. As unambiguous archaeological evidence of domestic reindeer origin and dispersal is unavailable, most theories have instead involved cultural historical models based on surveys of the geographic distributions of different styles of reindeer keeping, its associated equipment, and cultural and linguistic data. Among the early theories, there is the monocentric model of Laufer (1917) who argued that reindeer husbandry originated primarily in the Baĭkal area in southern Siberia at the beginning of our current era (CE [AD]), from where domestic reindeer spread to all other areas. Other scholars suggested polycentric models with more than one origin of initial domestication, implying that reindeer husbandry originated in the Sai͡ an and Baĭkal regions of Southern Siberia, as well as in Fennoscandia (Hatt, 1919;Maksimov, 1928;Wiklund, 1918). Recently, the I͡ Amal Peninsula (Fedorova, 2000) and the Polar Ural Mountains (Golovnëv, 1993) in Northwestern Siberia have been put forward as two further regions for the independent origin of reindeer husbandry. Many classic models of the diffusion of reindeer husbandry associate it also with the diffusion of domestication gear-halters, tethers, and saddles-from horses to reindeer, and sometimes from horses and dogs to reindeer (Vasilevich & Levin, 1951).
Convincing evidence implies that reindeer have been used by indigenous people for transport, as decoy animals to attract wild reindeer, and in some places for milking, probably long before large-scale pastoralism emerged. Rock art in Southern Siberia depicting reindeer used for riding and pulling sleds appears to date to the second millennium before CE (Okladnikov & Mazin, 1976). Archaeological remains of "boat-like coffins"-similar to the Sámi kind of sledge (gieres)-have been dated to the same period (Murashkin, Kolpakov, Shumkin, Khartanovich, & Moiseyev, 2016) and may suggest reindeer were used for transport for three millennia in parts of Northern Europe. In Northwestern Siberia, archaeological objects interpreted as reindeer headgear date as early as 100 before CE (Gusev, 2017), which some have argued indicate that reindeer were domesticated and used for transport by that time (Fedorova, 2000;Gusev et al., 2016). However, these fragmentary archaeological findings at best suggest the existence of a small-scale hunting-herding tradition, and not the widespread dependence on domestic animals as is associated with the pastoral tradition.
Many scholars assume that reindeer husbandry developed to provide a means of transport to facilitate wild game hunting. Domestic reindeer were probably only killed for sacrifice or when food from game was scarce-as is still the case with many small-scale forest herding societies (Bjørklund, 2013;Golovnëv, 1993;Ingold, 1988).
Over many centuries, this style of reindeer domestication resulted in no significant change in local modes of subsistence, which remained focused on hunting, fishing, and gathering. From the 17th to 18th centuries, the domestic herds of several Arctic people began to grow surprisingly quickly with subsequent emergence of a new form of economy known as pastoralism (Bjørklund, 2013;Ingold, 1988;Krupnik, 1976). The pastoralist transition is characterized by the use of reindeer in nearly all spheres of life starting from transport to a source of skins for clothing and to build dwellings, and as the primary source of food. Besides protecting and controlling their animals, the large herd sizes dictated that these reindeer herders were always on the move and adapted or re-engineered their earlier semi-sedentary inventions. As Stépanoff, Marchina, Fossier, and Bureau (2017) have pointed out the scale of the transition inverted the relation between people and reindeer such that the herders had to cater to the needs and the desires of the animals, whom they followed on migrations across the tundra.
The questions related to when, why, and how this pastoral transition took place have been the focus of recurrent scientific debates. In Scandinavia, many authors assume that the collapse in the numbers of wild reindeer from overhunting during 16th to 18th centuries, which accompanied the pressures of colonialism, the market economy, and increased demands for food, led people to change their subsistence strategies (Ingold, 1986;Lundmark, 2007;Vorren, 1973). Others point to long-term fluctuations in the climate during 17th, 18th, and 19th centuries, producing a favorable cooling cycle that facilitated the explosion of domestic reindeer populations (Krupnik, 1989;Stépanoff et al., 2017). Finally, both Scandinavian and Russian ethno-historians point to the need for Arctic indigenous groups to keep more animals on hand in order to meet tax and trading obligations with the expanding states around them (Krupnik, 1989;Vorren, 1978). Whatever the cause of the pastoral transition, scholars have speculated whether the rapid growth of domestic reindeer herds was facilitated by the import of a new domestic type (Bjørnstad, Flagstad, Hufthammer, & Røed, 2012;Røed, Bjørklund, & Olsen, 2018), or if wild local stocks were selectively captured and bred to generate a regional domestic type (Golovnëv, 1993).
The domestic reindeer studied are from nomadic herds managed by indigenous Nenetses (Figure 1). Nenetses inhabit the polar regions of Northwestern Siberia and Northeastern Europe and belong to one of several Samoedic linguistic groups whose traditional economy centered on reindeer husbandry (Stammler, 2005). Historical accounts of Samoeds in the 17th century suggest their subsistence centered on hunting and fishing and that their domestic herds were uniformly small scale and used primarily for transport or as decoys to attract wild migratory Rangifer (Krupnik, 1976). Beginning in the 18th century, the herd sizes of some families increased greatly, and many families began travel long distances with large domestic reindeer herds depending entirely on them for their dietary and material needs (Krupnik, 1976;Svoboda, Sázelová, Kosintsev, Jankovská, & Holub, 2011). Today Nenetses are the largest indigenous reindeer herding people of the Russian North, altogether herding around 1,000,000 animals (http://www.gks.ru/). In order to address the history of reindeer domestication in Northwestern Siberia, we have analyzed maternal DNA in ancient reindeer bones taken from archaeological sites with genealogies spanning from the Late Pleistocene to the 18th century, and from contemporary wild and domestic reindeer. Specifically, we tested for a temporal genetic change in reindeer associated with the pastoral transition in the region, determined when this change occurred, and whether the local wild stocks were used as basis to form the contemporary domestic breed.

| Archaeological material
Altogether 299 samples from bones, teeth, and antlers were ob- domestic reindeer (for more information and references to each site see Appendix S1). Site 15 is the youngest site, and dates to circa AD 1700, and therefore is marked separately in Figure 2. The site and bone assemblage chronology are based on the calibrated radiocarbon dating of 39 reindeer samples used in this study ( Figure S1 in Appendix S1) together with radiocarbon dating of samples from the artifact-bearing strata prior to this study (Hufthammer, Svendsen, & Pavlov, 2018;Losey et al., 2018;Hufthammer, Henriksen, and Pavlov, pers. com.). Samples from sites with archaeological stratification, and where radiocarbon dating supported stability, were separated into upper and lower excavation layers (Table 1; Appendix S1).
Suitable reindeer teeth, bones, or antlers were carefully selected, based on osteological determinations. When possible, different individuals were identified and classified according to morphology, size, and age group, to ensure that the samples within each site were from different animals.  Table S3 in Appendix S1).

| Laboratory methods
DNA isolation and amplification setup of the archaeological material were undertaken in spatially separated lab facilities with standard precautions for working with ancient samples (Hofreiter, Serre, Poinar, Kuch, & Paabo, 2001;Wandeler, Hoeck, & Keller, 2007). Each sample was cleaned, and the outer surface removed before being drilled to obtain approximately 10-20 mg of powder. DNA was extracted using DNeasy TM Tissue Kit (Qiagen) following the protocol described in Bjørnstad and Røed (2010 Table 1 and Appendix S1 for more information of samples and localities TA B L E 2 Pairwise genetic differences (F ST ) between reindeer obtained at various archaeological sites (sample codes 1-15) and sets of contemporary reindeer (sample codes 16-20) in Northwestern Siberia. The sample locations are as given in Table 1 Table S3 in Appendix S1).

| Data analyses
To obtain the maximum number of sequences for comparison, the amplicons were trimmed to 190 and 374 bp for the ancient and contemporary CR alignments, respectively. In this study, we compared the same 190 bp alignment in modern and ancient material. Arlequin v3.5 (Excoffier & Lischer, 2010) was used to estimate nucleotide and haplotype diversity and to obtain site pairwise F ST estimates based on haplotype frequencies. The statistical significance was evaluated using 1,000 permutations. Average numbers of nucleotide differences were calculated in DnaSP v5 (Librado & Rozas, 2009). The program SAMOVA v1.0 (Dupanloup, Schneider, & Excoffier, 2002) was used to define groups of sample sets that are spatially and temporally homogeneous and maximally genetically differentiated from each other. Differences in average nucleotide diversity measures between sample sets were assessed by analyses of variance (ANOVA) and Wald statistics using the lm-function in R (R Core Team, 2019).

| RE SULTS
Of the 299 ancient samples, 223 produced reproducible mtDNA sequences (Table S3 in Appendix S1). All 20 PCR products amplified for cloning gave identical sequences to these consensus sequences.
After cloning, seventeen cloned products gave a predominance of clones identical to the consensus sequence, and two products gave respectively two and one clone identical, while one product displayed all twelve clones analyzed to be different, none of which were identical to the consensus sequence. We analyzed altogether 199 clone sequences and identified a total of 243 nucleotide misincorporations mainly represented by C/G to T/A transitions (89%) together with some T/A to C/G transitions (9%) and few transversions (2%; Table S2 in Appendix S1). The dominate transitions are misincorporations typically seen when sequencing ancient DNA (Stiller et al., 2006). In all, both the repeated PCR and the dominance of clones identical to the consensus sequences give evidence of the consensus sequences to be little affected by the misincorporations and are thus considered to be mainly reliable.  Figure S1 in Appendix S1).
The low mean FST value among sample codes for these two sets, that is, 0.028 ± 0.023 and 0.038 ± 0.042, respectively, gives evidence of low genetic differentiation within these two sample sets.
On the contrary, the high value for the mean pairwise differences observed between the two sample sets ( Such a star-like pattern is thought to have been produced by an ancient population expansion with the most common haplotype assumed to be the ancestral haplotype (Slatkin & Hudson, 1991).
This ancestral haplotype dominated all modern domestic herds (frequency range 0.27-0.57) and the youngest archaeological site code 15 (frequency 0.38), but was absent in contemporary wild reindeer and present in only four of the older sites (one in each of sample code 3, 5, 10 and 13). Cluster II was represented by 15 haplotypes in 43 samples, of which all were found in contemporary wild reindeer (n = 6) or in the older sample codes 3-13 (n = 37). Subcluster Ib contained three haplotypes in 11 samples, two in modern domestic reindeer and nine in ancient reindeer (sample codes 3-12).

| The pastoral transition and the genetic shift
The high levels of genetic variation within a homogeneous genetic structure found within the wild gene pool both temporally and spatially suggest that relatively large populations of wild reindeer occupied the region since late Pleistocene. The significant genetic shift associated with the late-dated archaeological site Khali͡ ato-1 is striking and points toward a fundamental genetic change on the I͡ Amal that seems to have been completed by the start of the 18th century. This change coincides very roughly with the period associated with the onset of the pastoral tradition. Using historical sources, Krupnik (1989) associates the practice of holding of large herds of F I G U R E 3 Mean control region diversity and haplotype pairwise difference estimates, with ±1 SD, in reindeer across sample codes 1-14 (ancient), sample codes 15-19 (recent domestic) and sample code 20 (contemporary wild) in Northwestern Siberia domestic reindeer to the beginning of the 18th century, proceeding to a "transition" to pastoralism by the middle of the 19th century, and the establishment a fully socio-economic "revolution" by the end of the 19th century. Although we only have evidence from one archaeological site dating to this period, it seems reasonable to link the genetic shift to this pastoralist transition. According to our data, the genetic shift on I͡ Amal could have happened anywhere between the 15th and 18th centuries. With a greater variety of samples, it might be possible to date this genetic shift more precisely.
It should be noted that others have interpreted the many artifacts found at sites on the I͡ Amal Peninsula as signs of a social complexity and a remarkable inter-regional standardization of techniques built on large-scale reindeer transport and large-scale reindeer husbandry, beginning as early as 2,000 BP (Gusev et al., 2016).
If this was the case, according to our data, these processes must have involved domesticated animals that were genetically similar to the wild stock stretching back to the late Pleistocene.
The typical wild gene pool pattern which is dominant in all archaeological sites prior to the 15th century, suggests that people in the region relied primarily upon wild reindeer for food. However, it is possible that these hunting groups also held a small number of domestic reindeer to use as hunting decoys or for transport. The four samples presenting a typically modern domestic genetic signature (most common cluster Ie haplotype) at sites 3, 5, 10 and 13 could indicate that a small number of animals with a genetic constitution similar to those of the domestic reindeer today were present on I͡ Amal since as far back as 6,000 BP. If local wild reindeer were tamed (cf. Golovnëv, 1993), we would not be able to genetically distinguish them from the wild reindeer. If a different single animal was imported for a special purpose, for example to serve as decoy reindeer, it is likely that its signature would be rare. Likely these exotic animals were held in small numbers, were highly valued, and were not slaughtered for food (except in times of great starvation; Krupnik, 1993). Such sparse remains would naturally be lost or diluted among the larger amounts of bone material deposited from the remains of wild migratory reindeer. It is possible that the remains of domestic reindeer may have been curated differently to those of wild reindeer. We know that contemporary hunter-fisher Nenetses create offerings of domestic reindeer bones at sacred sites on the surface (Haakanson, 2000). Further, at the recent site Khali͡ ato-1, Rangifer remains were deposited on the surface as a ritual sculpture (Kardash & Sokolkov, 2016, cf. Appendix S1). Surface deposition would be poorly preserved across the millennium. What can be concluded is that the animals presenting cluster Ie features appeared first in a scattering of archaeological sites in small numbers, and then came to rapidly dominate the signatures of the domestic reindeer population of this region.
The contemporary domestic population also presents a limited number of haplotypes with reduced mean pairwise haplotype differences in the maternal gene pool. It also displays several new haplotypes not present in the old and modern wild gene pool. This suggests that the Nenets pastoral transition found its origins on a F I G U R E 4 Bayesian consensus tree (a) and network (b) for ancient and contemporary reindeer in Northwestern Siberia inferred using 190 bp of the control region. Bayesian posterior probabilities >.80 provided at the tree nodes. Clade labels used throughout the text are indicated as vertical bars beside the tree and as encircled haplotypes in the network. Each colored sphere represents unique haplotypes with the area proportional to the number of reindeer sharing a haplotype, and with colors representing the sample codes 1-14 (archaic) in red, sample code 15 (archaic) in green, sample codes 16-19 (modern domestic) in yellow and sample code 20 (contemporary wild) in blue limited number of individuals with a maternal ancestry comprising partly of non-native origin. It is possible that the rapid growth in herd sizes that has been documented from the 17th-19th centuries facilitated the development of a unique type based on small numbers of imported pioneers. Our data therefore associates the emergence of pastoralism with the actual translocation of a special type of animal, as has been also reported for pastoral transition in Fennoscandia (Røed et al., 2018). This strategy of importing specialized domestic stock is also seen for several other domesticated animal species (Clutton-Brock, 1999;Larson & Burger, 2013).
The pattern of temporal genetic change for the Nenets reindeer is very similar to what has been reported for reindeer in northern Fennoscandia, where reindeer appeared to have gone through a massive maternal mtDNA replacement during the 16th and 17th centuries, suggested to be associated to the onset of pastoralism (Bjørnstad et al., 2012;Røed et al., 2018). Here, the collapse in the numbers of wild reindeer, together with pressures from colonialism and market economy led people to keep more animals at hand (Ingold, 1988;Lundmark, 2007;Vorren, 1973). It is interesting to note that other reindeer herding peoples, such as Chukchis and Koriaks much further east in Russia also experienced a rapid growth in the size of their domestic herds in the 17th and 18th centuries (Krupnik, 1993).
The near-simultaneous emergence of pastoralism across the Eurasian Arctic which generally accompanied the decrease of the local wild reindeer (Ingold, 1986;Klein, 1980;Lundmark, 2007;Syroechkoveskii, 1995) suggests that there might have been some general drivers. This could be the onset of the Little Ice Age when most of the Arctic experienced the coldest sustained temperatures of the past 800 years, with the coldest interval occurring between the 17th and mid-19th centuries (Briffa et al., 2013;Kaufman et al., 2009). Rangifer respond well to damp cool summers and deeply cold winters (Weladji, Klein, Holand, & Mysterud, 2002). Increasing domestic herd sizes perhaps allowed for increased human mobility that in turn greatly facilitated the hunting of wild reindeer (Krupnik, 1993;Stépanoff, 2017).
Consequently, at the turn of 19th century, the decline of the wild reindeer population did not precede but rather followed the increase in domestic herds. According to this scenario, the main facilitating factor of the transition from hunting to pastoralism was climate, although local political and economic factors may have been influential as they incited herders to keep large herds. Following the initial transition to pastoralism when people shifted toward subsisting primarily on domestic animals, further reduction and eventually depletion of the wild reindeer may have taken place due to the challenges of the coexistence of large wild and domestic herds (Baskin, 2005;Klein, 1980).

| The genetic structure and ancestry of I͡ Amal-Nenets domestic reindeer
The contemporary domestic breed displays several CR haplotypes not present in the wild gene pool. Admittedly, the data show a large number of haplotype singletons implying that more haplotypes might be discovered with a greater variety of archaeological cases.
Most private domestic haplotypes were closely related to a few common haplotypes suggesting that the domestic breed was built from a few main maternal lineages. The fact that domestic herds exploded in size over the last few centuries enhance the probability that some of these unique private haplotypes represent mutations in the hypervariable part of the CR over this time span. However, the high proportion of private domestic haplotypes (i.e., about 50%), and the small fraction haplotypes belonging to the domestic lineages among the archaeologic sites 1-14, nevertheless suggest that the dominant domestic type of today had its origins in animals imported from other areas.
Several mechanisms could have been involved in introducing new maternal genomes to the region. This polar region experiences periodic crashes in the number of reindeer due to poor weather or spring icing events (Briffa et al., 2013;Forbes et al., 2016;Klokov, 2011).
Both historical sources and oral history document significant collapses of reindeer due to diseases or icing events in 1911 and 1922 (Podkorytov, 1995). It is therefore conceivable that subsequent to previous population crashes the herds were replenished from other reindeer-herding regions to the south or to the east. There are also documented state-led experiments in translation during the Sovietera of importing domestic reindeer from Chukotka to improve local domestic herds (Podkorytov, 1995). Furthermore, low genetic differentiation characterizes the domestic reindeer herds analyzed in this study. Both the long-term pattern of reindeer transfer due to dowry and friendship exchanges and later state-led breeding and exchange of particular animals may have contributed to the development of a distinct and relatively homogeneous genetic structure for the Nenets domestic reindeer breed seen today.
If we reflect on the fact that the same homogeneous genetic signature is present from as late as the 18th century site on I͡ Amal, it would seem that the genetic shift must have predated the era of Soviet zootechnics. Further, the fact that the mean haplotype differences are low among the domestic reindeer point toward a common evolutionary history characterizing the domestic maternal gene pool.
Most private haplotypes in the domestic population were within the CR cluster with the signature of ancient population expansion (Ie, Figure 4b). This, together with the dominance of domestic animals among also the ancestral haplotype of this cluster, may indicate a domestic ancestry in association with the distribution and history of this lineage. Subcluster Ie haplotypes are commonly distributed in other domestic herds within both NAD, KR and Kola Peninsula (Kvie, Heggenes, Anderson, et al., 2016;Kvie, Heggenes, & Røed, 2016;Røed et al., 2008). The near absence of this lineage among modern wild reindeer, as detected here for the large Taĭmyr population, has been also reported for the Belyĭ Island population located directly north of the I͡ Amal Peninsula, and also for the Novai͡ a Zemli͡ a population located north and west of the I͡ ANAD (Kvie, Heggenes, Anderson, et al., 2016). The fact that this maternal lineage is not found in contemporary northern wild reindeer populations may suggest that the contemporary Nenets domestic breed has roots toward the south or southwest.
While the diverse CR cluster I is suggested to represent a Euro-Beringian lineage evolved from a Pleistocene population in northern Eurasia (Flagstad & Røed, 2003;Yannic et al., 2014), the evolutionary history of its subclusters is more uncertain. The relatively shallow divergence of Ie from other cluster I subclusters implies a more recent divergence from the general Euro-Beringia lineage as also reported for estimates of time since most recent common ancestor for the different cluster I subclusters (Kvie, Heggenes, & Røed, 2016). As outlined above, one hypothesis is that small numbers of an exotic animal could have been early imported, were selectively bred beside other habituated local reindeer, and with the help of mutations amplified through an exploding population, their signatures came to dominate the population of domestic reindeer we know today. It is also possible that a strong geophysical event, possibly associated to the Holocene warm period (9,000-5,000 BP, Kaufman et al., 2004), may have isolated this population away from the general Euro-Beringia lineage, followed by a sudden expansion of the population due to more favorable climate, could explain the history of this lineage. Possible refugees could have been in the Ural Mountains (Salonen et al., 2011) or the Sai͡ an Mountains of Southern Siberia (Laufer, 1917;Prokov'ev, 1940).
Although our results may support a Southern Siberian ancestry to the Nenets reindeer breed, which might be just one of several areas from which reindeer herding and domesticated reindeer spread northwards, there is a need for more data before being able to specify the origin point for the non-native ancestry of this breed.
Besides, the use of other genetic markers may imply a more nuanced Bjørklund and Bjørnar J. Olsen for useful comments on an earlier version of this paper. We are especially grateful to two anonymous reviewers for their comments, which greatly helped to develop the manuscript.

CO N FLI C T O F I NTE R E S T
We declare no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data for this study will be available in the Sequence Database at the National Centre for Biotechnical Information (NCBI; GenBank ID: MT146049-MT146444).