Genetic Diversity, Admixture and Analysis of Homozygous-by-Descent (HBD) Segments of Russian Wild Boar

Simple Summary The wild boar is one of the most common wild animals. On the territory of Russia, there are two subspecies of the wild boar—European and Asian. At the beginning of the 20th century, the wild boar in the European part of Russia was practically exterminated. Later the population was restored by importing animals from other regions and by self-repopulation. The aim of our research was to assess the population structure of the Russian wild boar in comparison with the wild boar from other regions of the world and to determine the level of autozygosity, which allows us to determine the state of the population. We found traces of introgression of the Asian wild boar into the European one due to, migration of the wild boar in the population recovery process. Further analysis for genetic influx into Russian wild boar population identified four samples in which more than 10% of the genome belonged to domestic pigs. The Homozygous-by-Descent (HBD) Segments evaluation showed a low level of autozygosity in comparison with the aggregate sample of the European wild boar. Based on our genetic evaluation, we concluded that the population of the Russian wild boar of the European and Asian subspecies are characterized by a sufficient level of genetic diversity. Abstract The wild boar is the wild ancestor of the domestic pig and one of the most common species of ungulates. At the beginning of the 20th century, the wild boar was practically exterminated in the European part of Russia. In the period 1935–1988, 7705 boars were caught in various regions of the European part of Russia, the Far East, Ukraine, Belarus, Kyrgyzstan, Kazakhstan, Latvia, Lithuania, Estonia, Tajikistan and resettled in the territory of Russia. Asian and European wild boars dwell the territory of Russia. The aim of our research was to study the genetic diversity and structure of wild boar populations in different regions of Russia using genome-wide genotyping. We have determined the genetic distances, population structure, parameters of genetic diversity and significantly expanded our understanding of the genetic state of the Russian wild boar. For the first time, we calculated autozygosity of the wild boar of the European and Asian subspecies using Homozygous-by-Descent (HBD) Segments analysis, which is important in terms of population recovery. We also found evidence of hybridization between Russian wild boar and domestic pigs. A group of European wild boars showed introgression of the Asian boar into population. The mean level of the inbreeding coefficient in European wild boar was higher than in Asian wild boar, and combined groups of the European boar had higher inbreeding coefficient than Russian wild boars. These results obtained can be used in population management.

Assessment of genetic diversity and population structure is a fundamental task. This is important for understanding the evolutionary history of its origin, and to provide important information for the conservation and management of biodiversity [25,26]. In this aspect, the assessment of the wild boar is of particular interest, since both of its subspecies-Asian and European wild boars live on the territory of Russia. However, the origin and genetic structure of the Russian population has not been sufficiently studied. In this regard, the aim of our research was to study the genetic diversity and structure of wild boar populations in different regions of Russia using genome-wide genotyping.
In order to increase the accuracy of SNP genotyping during the quality check, GenCall (GC) and GenTrain (GT) scores were used. GC and GT scores of 0.5 cut-off were applied to determine the valid genotypes for each SNP [31].
The files were merged according to the standard bcftools -merge algorithm with the removal of duplicated SNPs on the same positions. We collected several datasets for complex processing. The first dataset included samples from public databases of wild boars from different parts of the world and those of Russian animals. The second dataset was created to exclude introgression between commercial pig breeds and Russian wild boars. The third dataset included only animals selected on the territory of Russia. The fourth dataset was separated on the basis of ancestry and included four groups: RU_EU and RU_AS animals inhabiting the European and Asian parts of the Russian Federation, respectively, WB_EU and WB_AS animals from different European and Asian regions of the world.
We recommend to select only autosomal chromosomes 1-22, via the PLINK "autosome" option [32]. This option avoids detecting structures that are gender-biased. Filtering genotypes with missing rate > 20% was conducted in PLINK with option "-geno 0.2". SNPs with high rate of missingness and low minor allele frequency (MAF < 0.01) were removed through the PLINK option "-maf 0.01". Too rare SNPs (MAF < 0.01) could be found at an individual level but are not commonly present at a population level. Samples with a call rate at least 70% by specifying "-mind 0.3" in PLINK were allowed for analysis. SNP pruning was conducted with filtering out SNPs in linkage disequilibrium (LD) blocks using the PLINK option "-indep-pairwise 50 5 0.2". We assume low or no correlation structure between SNPs as suggested via r2 < 0.2, with r2 the commonly used measure of LD [33]. LD pruning in this way helps to avoid that strong LD blocks drives the most important calculations, for example-Principal Components or Admixture analysis [34]. Hardy-Weinberg equilibrium test was not performed for comparisons with worldwide population because of Wahlund effect [35]. Upon the completion of filtering, the global and Russian wild boar dataset included 14,491 SNPs (579 samples with 96% genotyping quality), the commercial breed of pig and Russian wild boar-9490 (442 sample with 97% genotyping quality). Homozygous-by-Descent (HBD) Segments of combined groups of European wild boars (WB_EU, n = 387) and Asian wild boars (WB_AS, n = 34) were compared to Russian European (RU_EU, n = 94) and Russian Asian wild boars (RU_AS, n = 64).

Genetic Distances and Population Structure
We used PLINK v1.09 to conduct Principal Component Analysis and to visualize the results of comparison inside the selected groups and plotted a scatterplot using the R language. The calculation of Ar (allelic richness); Ho (observed heterozygosity); He (expected heterozygosity); Fis (inbreeding coefficient) were performed using package diversity in R [36]. Average minor allele frequency for each population was calculated with PLINK 1.9. The rate variation among sites was modelled with a gamma distribution (shape parameter = 1) [37]. We applied unsupervised hierarchical clustering of individuals using the maximum likelihood method implemented in ADMIXTURE v.1.3. We used default input parameters for estimation of the admixture components [37]. In addition, '-cv' flag was enabled to perform the cross-validation procedure and to calculate the optimal k value for all populations.

Analysis of Homozygous-by-Descent (HBD) Segments
Realized inbreeding coefficients were calculated using the R package RZooRoH v.0.3.0.74 [38]. This package is based on the hidden Markov model that identifies Homozygousby-Descent (HBD; evident from ROH) and non-HBD segments. This allows evaluation of the current level of inbreeding better by assuming inbreeding as genome-wide and as local scales, and classifying HBD segments into age-based classes [39]. RZooRoH allows to determine approximate generation classes based on the length of the segments. The different HBD classes are defined by their specific rates Rk. The length of HBD segments from class k is exponentially distributed with the rate Rk and mean 1/Rk. Classes with lower rates correspond to longer HBD segments from more recent common ancestors. Therefore, different HBD classes can be interpreted as HBD segments of different groups of ancestors tracing back to different generations in the past. The rate of the class is approximately equal to the size of the inbreeding loop in generations [39].

Genetic Diversity and Admixture of the Russian Wild Boar
We assessed the position of the Russian wild boar in the Large Scale Genetic Structure of wild boar from different regions of the world ( Figure 2).
guage. The calculation of Ar (allelic richness); Ho (observed heterozygosity); He (expected heterozygosity); Fis (inbreeding coefficient) were performed using package diversity in R [36]. Average minor allele frequency for each population was calculated with PLINK 1.9. The rate variation among sites was modelled with a gamma distribution (shape parameter = 1) [37]. We applied unsupervised hierarchical clustering of individuals using the maximum likelihood method implemented in ADMIXTURE v.1.3. We used default input parameters for estimation of the admixture components [37]. In addition, '--cv' flag was enabled to perform the cross-validation procedure and to calculate the optimal k value for all populations.

Analysis of Homozygous-by-Descent (HBD) Segments
Realized inbreeding coefficients were calculated using the R package RZooRoH v.0.3.0.74 [38]. This package is based on the hidden Markov model that identifies Homozygous-by-Descent (HBD; evident from ROH) and non-HBD segments. This allows evaluation of the current level of inbreeding better by assuming inbreeding as genome-wide and as local scales, and classifying HBD segments into age-based classes [39]. RZooRoH allows to determine approximate generation classes based on the length of the segments. The different HBD classes are defined by their specific rates Rk. The length of HBD segments from class k is exponentially distributed with the rate Rk and mean 1/Rk. Classes with lower rates correspond to longer HBD segments from more recent common ancestors. Therefore, different HBD classes can be interpreted as HBD segments of different groups of ancestors tracing back to different generations in the past. The rate of the class is approximately equal to the size of the inbreeding loop in generations [39].

Genetic Diversity and Admixture of the Russian Wild Boar
We assessed the position of the Russian wild boar in the Large Scale Genetic Structure of wild boar from different regions of the world ( As the scatterplot shows, components 1 and 2, as well as 1 and 3, separate the European and Asian boar. The European wild boar from southern regions is distanced from more northern regions, with the exception of a few samples of the SAR population. It is noteworthy that the Russian European wild boar is located in close proximity to the Finnish wild boar and the wild boar of the countries of Eastern Europe. The Russian Asiatic wild boar forms a single group with the Chinese wild boar; wild boars from Thailand form a separate subgroup. The overall analysis of the population shows a clear division into European and Asian boars at K = 2 ( Figure 3A). RUEU boars have traces of the introgression of the Asian boar. At K = 28, both RUAS and RUEU groups of wild boars shows 3 clusters. The mini- As the scatterplot shows, components 1 and 2, as well as 1 and 3, separate the European and Asian boar. The European wild boar from southern regions is distanced from more northern regions, with the exception of a few samples of the SAR population. It is noteworthy that the Russian European wild boar is located in close proximity to the Finnish The overall analysis of the population shows a clear division into European and Asian boars at K = 2 ( Figure 3A). RUEU boars have traces of the introgression of the Asian boar. At K = 28, both RUAS and RUEU groups of wild boars shows 3 clusters. The minimum Cross-validation error was for K = 18 ( Figure S1, Table S2). We performed an admixture analysis identifying potential hybridization between the Russian wild boar and domestic pigs. Estimation of the parameter Q at K = 5 for which the minimum Cross-validation error was calculated, showed evidence of hybridization between Russian wild boar and domestic pigs for 53 samples of European wild boars. The total share between pig and boar genome was 0.010-0.568 ( Figure S2, Table S3). The largest portion of the domestic pig genome was found in a wild boar from the Leningrad region, which was apparently a hybrid of the first generation (0.117 and 0.057 belonged to two clusters of Duroc pigs), and more than 0.100 domestic pig genome was found in four wild boars from Krasnodar.

Analysis of Homozygous-by-Descent (HBD) Segments
The combined groups of European wild boars (WB_EU), Asian wild boars (WB_AS) were compared to Russian European (RU_EU) and Russian Asian wild boars (RU_AS). The mean level of the inbreeding coefficient of the European wild boar was higher than that of the Asian wild boar, and the combined groups of the European boar had a higher inbreeding coefficient than those of the Russian wild boar (Table 2). The highest level of total autozygosity was found in the WB_EU group and amounted to 0.168 and the lowest in the RU_AS group was 0.025.

Figures 4 and 5 demonstrate the contribution of different HBD segments to total autozygosity.
THAI and TUN. The average within-population MAF varied from 0.212 (RUAS) to 0.294 (SER) ( Table 1).

Analysis of Homozygous-by-Descent (HBD) Segments
The combined groups of European wild boars (WB_EU), Asian wild boars (WB_AS) were compared to Russian European (RU_EU) and Russian Asian wild boars (RU_AS). The mean level of the inbreeding coefficient of the European wild boar was higher than that of the Asian wild boar, and the combined groups of the European boar had a higher inbreeding coefficient than those of the Russian wild boar (Table 2). The highest level of total autozygosity was found in the WB_EU group and amounted to 0.168 and the lowest in the RU_AS group was 0.025.  made by the middle segments R_8-R_64. In the RU_AS group, although the total share of segments was low the segments of R_64-R_256 classes were prevalent. The Rk512 coefficient in the WB_EU group was 0.053; in the WB_AS group Rk512 it was 0.025 (Figures 3 and 4). In the Russian wild boars, it was determined at 0.005-0.009. WB_EU group of wild boars showed higher level of Group inbreeding coefficients by HBD-classes ( Figure 4) whereas in other groups the inbreeding was at similar level.  Table 3 presents the results of the number and lengths of segments analysis in groups.  In the WB_EU group, the short R_512 segments, which show the presence of common ancestors about 250 generations ago, made a greater contribution to autozigosity. In addition, the long and medium R_8-R_64 segments made a rather large contribution. In the WB_AS group, a large share was occupied by long segments R_4, medium segments R_64 and short segments R_256-R_512. In the EU_RU group, there were practically no short segments R_256 and R_512, while larger contribution was made by the middle segments R_8-R_64. In the RU_AS group, although the total share of segments was low the segments of R_64-R_256 classes were prevalent.
The Rk 512 coefficient in the WB_EU group was 0.053; in the WB_AS group Rk 512 it was 0.025 (Figures 3 and 4). In the Russian wild boars, it was determined at 0.005-0.009. WB_EU group of wild boars showed higher level of Group inbreeding coefficients by HBD-classes ( Figure 4) whereas in other groups the inbreeding was at similar level. Table 3 presents the results of the number and lengths of segments analysis in groups. Samples of the WB_EU group are characterized by a larger number of segments compared to other groups (Table 3). In all groups, except for WB_AS, the largest number of detected segments belonged to the Rk 512 class-22.8-28.5%.
We examined the distribution of segments depending on the class and their location on the chromosomes ( Figure 6).
Samples of the WB_EU group are characterized by a larger number of segments compared to other groups (Table 3). In all groups, except for WB_AS, the largest number of detected segments belonged to the Rk512 class-22.8-28.5%.
We examined the distribution of segments depending on the class and their location on the chromosomes ( Figure 6). The profiles of segment distribution of different classes on chromosomes between the groups differed. Long segments were localized mainly on the 1st, 2nd, 3rd, 4th, 6th chromosomes in the Russian wild boars while there were no segments of the Rk2 class on the 1st chromosome in the WB_AS group, but a large proportion of them was recorded on the 2d chromosome, as well as in the RU_AS group. In the WB_EU group, segments of all studied classes were found on all chromosomes.

Discussion
We investigated the genetic profile of the Russian wild boars, which currently remains poorly studied. There are two subspecies of wild boar that inhabit Russia. The European boars live in the European part of Russia, the Asian boars live in the Primorsky The profiles of segment distribution of different classes on chromosomes between the groups differed. Long segments were localized mainly on the 1st, 2nd, 3rd, 4th, 6th chromosomes in the Russian wild boars while there were no segments of the Rk2 class on the 1st chromosome in the WB_AS group, but a large proportion of them was recorded on the 2d chromosome, as well as in the RU_AS group. In the WB_EU group, segments of all studied classes were found on all chromosomes.

Discussion
We investigated the genetic profile of the Russian wild boars, which currently remains poorly studied. There are two subspecies of wild boar that inhabit Russia. The European boars live in the European part of Russia, the Asian boars live in the Primorsky Territory, Amur, Irkutsk Region and Khabarovsk. Our data confirmed (Figures 1 and 2) that European and Asian wild boars differ from each other as indicated by other researchers [20,40,41]. Analysis of genetic profile showed that the European subspecies of the Russian wild boar is characterized by the presence of traces of introgression of the Asian wild boar. This can be explained by the settling of wild boar from the Far East and Primorsky Krai in the territory of the Moscow region , Tver region , Vladimir region , Kaluga region (1964)(1965)(1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974)(1975)(1976)(1977)(1978)(1979)(1980)(1981), Yaroslavl region , Nizhny Novgorod region (1963)(1964)(1965)(1966)(1967)(1968)(1969), Volgograd region (1969), Sverdlovsk region (1978)(1979)(1980)(1981)(1982)(1983)(1984) [6]. As shown in Table 1, the parameters of genetic diversity in the European wild boar groups are higher than in the Asian wild boar groups. Minor Allele Frequency (MAF), is an indicator of the abundance of rare alleles through the genome. In our study it ranged from 0.212 to 0.294 in our study, for Spanish was wild boars-0.192 [42], for different populations of European wild boar was in the range of 0.136-0.174 [43], the average MAF within local breeds ranged from 0.133 to 0.294 [42,43], for domestic pig of Landrace and Large White was breeds higher than 0.250 [44]. Xiao et al. showed that a low MAF can indicate a vulnerable position of population while high MAF in a small population can be a sign of high tolerance for inbred depression [45]. We found negative inbreeding coefficient (Fis) values for wild boar groups from Belgium, Bulgaria, France, Luxembourg, Serbia, Thailand and Tunisia, apparently this is due to the small size of the groups. Sánchez-Montes et al. pointed out the influence of population size and the presence of close relatives on the parameters of genetic diversity [46]. Analysis of the genetic diversity of the Russian European wild boar as compared with world-wide population showed the highest rate of allelic diversity, heterozygosity, while the genetic diversity of the Russian Asian wild boar is reduced. Slightly higher genetic diversity of the Russian European boar were indicated by Ho and He data that were for RUAS 0.185 and 0.190, for RUEU 0.328 and 0.342, for the wild boar from Iran 0.284 and 0.249; for the wild boar from Eastern Europe 0.325 and 0.312, for the wild boar from Western Europe 0.301 and 0.296 [47]. The study of the genetic diversity of the Asian wild boar using microsatellites showed that the allelic diversity and the level of heterozygosity of the Russian Asian wild boar were lower than those of the Chinese wild boar [16]. Analysis of a polymorphism of mtDNA showed that the level of genetic diversity of wild boars is higher in East Asia than in Europe [48].
Iacolina et al. [43] did not find hybrids in the Central-North-Eastern Europe wild boar population and also among four sample of Russian wild boars, but many researchers indicate traces of introgression of wild boar and domestic pigs [43,[49][50][51]. We analyzed the Russian wild boars and commercial breeds of pigs and did not find any traces of hybridization with Asian wild boars, but we found strong traces of such hybridization for one sample from Leningrad Region and less strong for 4 boars from Krasnodar. Goedbloed et al. [51] found for geographically widespread presence of domestic pig SNPs in 10% of analyzed wild boars of Northwest Europe. The lack of hybrids among the Asian wild boars can be explained by the more severe environmental conditions that are not favorable for pig breeding. On the territory of the European part of Russia, there are more favorable conditions for non-industrial pig breeding; accordingly, there is a high probability of hybridization of wild boar with domestic pigs.
Since the wild boar in the European part of Russia was practically exterminated and restored and expanded its habitat since the second half of the 20th century, it is important to understand the level of autozygosity in the modern wild boar population [6]. We examined HBD segments to assess the autozygosity of the Russian European and Russian Asian wild boar. The level of total autozygosity in Russian populations of wild boar was lower than in other groups. When compared with domestic pigs, the level of autozygosity of wild boar was lower. The studies of Landrace and Large White pig showed that total autozygosity (inbreeding coefficient) was approximately 0.23 (0.15-0.34) [52], for different populations of wild boar was in the range of 0.025-0.168. Short segments are practically absent in Russian wild boar. This can be explained by the relatively recent recovery of the modern wild boar population in Russia. The total number of segments in domestic pigs is also significantly higher, approximately 30,000-50,000, although more than 28,000 of segments have been identified in the European wild boar group (73.56 segments per an individual), but no more than 2000 segments have been identified in other wild boar groups (5.33-55.71 segments per an individual). Our studies have shown that HBD segments are evenly distributed across chromosomes in WB_EU, as previously reported by Bosse et al. [53], which also indicates that Chinese boars represent the most variable cluster due to their high nucleotide diversity and low amount of ROH (p < 0.001). In our case, the Asian boars had significantly fewer HBD segments than the European boars.
Our study investigated a genetic profile of the Russian wild boar population, determined the level of autozygosity, found the introgression of the domestic pig genome into European Russian boar and showed that the Asian Russian boar population is in a vulnerable position due to reduced genetic diversity.

Conclusions
Our study provides a comparative analysis of the genetic diversity of wild boars of various origins. Using various methods of processing genome-wide genotyping data, it was demonstrated that wild boars from different regions of the world differ in their genetic structure and the level of autozygosity. We found traces of domestic pig introgression in the Russian European wild boar. Our results can be useful for characterizing different populations of wild boar and domestic pigs. They provide important information for the conservation and management of biodiversity and expand understanding of the genetic state of the Russian wild boar population.

Data Availability Statement:
The raw data supporting the conclusions of this article will be made available by the authors upon reasonable request.