Skip to main content
Download PDF

Genetic diversity and signatures of selection in Icelandic horses and Exmoor ponies

BMC Genomics volume 25, Article number: 772 (2024) Cite this article

Abstract

Background

The Icelandic horse and Exmoor pony are ancient, native breeds, adapted to harsh environmental conditions and they have both undergone severe historic bottlenecks. However, in modern days, the selection pressures on these breeds differ substantially. The aim of this study was to assess genetic diversity in both breeds through expected (HE) and observed heterozygosity (HO) and effective population size (Ne). Furthermore, we aimed to identify runs of homozygosity (ROH) to estimate and compare genomic inbreeding and signatures of selection in the breeds.

Results

HO was estimated at 0.34 and 0.33 in the Icelandic horse and Exmoor pony, respectively, aligning closely with HE of 0.34 for both breeds. Based on genomic data, the Ne for the last generation was calculated to be 125 individuals for Icelandic horses and 42 for Exmoor ponies. Genomic inbreeding coefficient (FROH) ranged from 0.08 to 0.20 for the Icelandic horse and 0.12 to 0.27 for the Exmoor pony, with the majority of inbreeding attributed to short ROHs in both breeds. Several ROH islands associated with performance were identified in the Icelandic horse, featuring target genes such as DMRT3, DOCK8, EDNRB, SLAIN1, and NEURL1. Shared ROH islands between both breeds were linked to metabolic processes (FOXO1), body size, and the immune system (CYRIB), while private ROH islands in Exmoor ponies were associated with coat colours (ASIP, TBX3, OCA2), immune system (LYG1, LYG2), and fertility (TEX14, SPO11, ADAM20).

Conclusions

Evaluations of genetic diversity and inbreeding reveal insights into the evolutionary trajectories of both breeds, highlighting the consequences of population bottlenecks. While the genetic diversity in the Icelandic horse is acceptable, a critically low genetic diversity was estimated for the Exmoor pony, which requires further validation. Identified signatures of selection highlight the differences in the use of the two breeds as well as their adaptive trait similarities. The results provide insight into genomic regions under selection pressure in a gaited performance horse breed and various adaptive traits in small-sized native horse breeds. This understanding contributes to preserving genetic diversity and population health in these equine populations.

Peer Review reports

Background

Monitoring the genetic diversity within populations is vital to ensure sustainable breeding and should be performed routinely within breeding programs [1, 2]. This especially applies to breeding programs involving closed populations, such as the Icelandic horse and the Exmoor pony breeds. Both breeds are ancient, native breeds adapted to harsh environmental conditions, and they have both undergone severe historic bottlenecks, albeit more pronounced in the Exmoor pony breed. However, in modern days, the selection pressures on these breeds differ substantially. The Icelandic horse has been bred primarily for its performance in five gaits, while conservation efforts have been the focus for the endangered Exmoor pony. Despite their similar starting points, the divergent breeding goals offer a unique opportunity to study the effects of artificial and natural selection by comparing the genomic selection signatures in these two breeds.

Little is known with certainty about the origin of the Icelandic horse breed, but it is generally believed to have descended from horses brought to the country by Norse settlers around 1100 years ago [3]. Since the settlement, the horses have remained isolated in Iceland and survived harsh weather conditions and natural disasters, such as volcanic eruptions, without significant introduction of foreign genetic material [3, 4]. Before selective breeding started in the 20th century, the horse was mainly used for labour and transportation and was primarily shaped by its harsh natural habitat. In the 1950s, the first official breeding goal, emphasising a versatile riding horse with five gaits, was introduced, resulting in a shift in the selection criteria for the breed [4]. Selective breeding became prevalent, and already in the 1980s, the official breeding program adopted the method of best linear unbiased prediction (BLUP) animal model to estimate breeding values [5, 6]. In the wake of selective breeding and the increased global popularity of the breed during the late 20th century, particularly in northern Europe, the population size surged. In 1959, the population counted approximately 30,000 horses [7], but to date, approximately 300,000 horses are registered across 31 countries [8].

The Exmoor pony, much like the Icelandic horse, is an ancient native breed adapted to harsh conditions. This breed has coexisted with humans in Britain for over 30,000 years. A stud book for the Exmoor pony was established in 1921 to promote the breeding of purebred Exmoor ponies and ensure they retain the traits and characteristics of their ancestors [9]. However, the breed faced a severe population bottleneck during World War II, dwindling to about 50 individuals. Consequently, conservation efforts have prioritized the Exmoor pony, implementing a breeding program specifically designed for its preservation [9]. The breed is named after the high moorland in north-western Somerset and northern Devon, England, where these ponies traditionally roam free. However, Exmoor ponies are also bred at other sites in the UK, Europe, and North America. Today, there are approximately 500 ponies on Exmoor and an additional 3500 Exmoor ponies in various locations across the UK and other countries [9]. About 500 breeding mares and 100 licensed, registered stallions globally produce between 100 and 150 foals annually. Each foal born to registered parents is inspected by trained inspectors to ensure that the Exmoor pony’s characteristics and traits are maintained. The Exmoor pony breeding, therefore, focuses on maintaining breed standards, particularly regarding exterior features like coat colour and conformation [9].

Natural and artificial selection tends to reduce genetic variability within targeted genomic regions, resulting in increased homozygosity. These so-called signatures of selection in the genome can be studied using modern genomic methods, such as estimations of continuous homozygous segments called runs of homozygosity (ROH) [10, 11]. To date, estimates of ROH have been used to identify genomic regions potentially under artificial selection in multiple horse breeds. Several genomic regions associated with selection for athletic performance have been identified [12,13,14,15,16], and previously documented target gene (DMRT3) related to gait pattern has been confirmed [17, 18]. Furthermore, genomic regions associated with selection for complex traits such as temperament, disease susceptibility, and fertility have been suggested [12, 13, 17,18,19,20,21] as well as regions associated with coat pigmentation characteristics and morphological traits such as body size [17, 18, 21,22,23].

ROH can be caused by the mating of related animals and are, therefore, a measure of inbreeding [10, 11]. In general, short ROHs indicate distant inbreeding, but longer ROHs (> 5.0 Mb) suggest more recent inbreeding where the common ancestor occurs approximately up to 10 generations back [10]. The genomic inbreeding coefficient FROH is defined as the proportion of the autosomal genome that lies within ROH above a specified length [24]. Recent studies on different horse breeds have reported FROH estimates to range from 0.10 to 0.29 in breeds with closed stud books [14, 18,19,20,21,22,23, 25, 26]. In contrast, much lower coefficients have been estimated in breeds with semi-open stud books, such as the Swedish Warmblood horse (FROH = 0.006) [12].

A recent estimate of the mean pedigree-based inbreeding coefficient (FPED) for all Icelandic horses born in Iceland 2020, was reported to be 0.03 [27]. The effective population size (Ne) for the same cohort was estimated to range from 95 to 103 horses depending on the pedigree completeness index [27]. Inbreeding coefficients for the Icelandic horse population have also been estimated using genomic data. An estimate of the average genomic inbreeding coefficient based on microsatellite data was 0.04 [7], while those based on medium-density single-nucleotide polymorphism (SNP) data ranged from 0.08 to 0.13 [16, 18, 28,29,30]. For the Exmoor ponies, estimated genomic inbreeding coefficients have been reported to range from 0.17 to 0.25 [12, 18, 29].

Genomic data has furthermore been used to estimate the effective population size of both the Icelandic horse and Exmoor pony breeds. For the Icelandic horse, the Ne estimates varied depending on the type of genomic data used: ranging from 215 individuals based on microsatellite data from 442 horses [7] to 555 individuals based on SNP array data from 25 horses [29]. For the Exmoor pony, the Ne was estimated at 216 individuals based on a sample of 24 ponies with SNP array data [29]. Additionally, studies using medium-density SNP array data identified signatures of selection on equine chromosomes (ECA) 3, 10, 11, 15, and 23 in the Icelandic horse [18, 31, 32]. In contrast, a larger number of ROH islands were identified on ECA1-4, 6, 9, 11, 16, 18–19, 22–23, 28, and 30 in Exmoor ponies [17, 18, 31].

Due to inconsistencies in estimates between previous studies regarding especially genetic diversity in the Icelandic horse, as well as indications from pedigree analysis of a decreasing effective population size, updated estimations for this breed based on a larger data set and high-density SNP information are desired. The comparison with the Exmoor pony gives a valuable opportunity to distinguish between detected signatures of selection for performance, and signatures resulting from adaptations to harsh environment.

The aim of this study was therefore to assess genetic diversity and identify runs of homozygosity in the two breeds, and to estimate and compare genomic inbreeding and signatures of selection. We hypothesized that these breeds would share some signatures of natural selection for adaptation in their genomes, whereas signatures of artificial selection for performance would be specific for the Icelandic horse.

Methods

Sample collection

The study included 380 privately owned Icelandic horses born between 1993 and 2016, of which 166 were stallions or geldings and 214 were mares. Hair samples were collected from the horses’ tails, and the collection was performed at breeding field tests and visits to trainers and breeders in Iceland and Sweden. The majority of horses were born in Iceland (N = 299) and Sweden (N = 72), while a few were born in Denmark, Germany, and Norway (N = 9). According to previous studies Icelandic horses are well genetically connected within Iceland [7] and across country borders in continental Europe [33], indicating a comparable genetic background of horses in the sample. The sampled individuals were originally selected for different genome-wide association studies; half of the individuals were selected based on mane growth characteristics [34], while the other half was randomly chosen at breeding field tests [35]. All but ten geldings had been shown at a breeding field test and are therefore a part of preselected Icelandic horses more likely to contribute genetic material to future generations [36]. Based on pedigree data from the international Worldfengur database [8], the closest relatedness observed between individuals in the sample were two parent-offspring pairs. In addition, less than 1% of all possible relationships in the dataset were closer than half sibs but less related than full sibs. Efforts were made to balance the contributions from different families and avoid stratification in the data.

Genotype data for 280 Exmoor ponies was retrieved from a previous publication where details concerning data collection are described [37]. The Exmoor ponies were originally selected based on their insect bite hypersensitivity status, avoiding close relatedness as far as possible based on a complete pedigree data four generations back. Furthermore, three subpopulations were reported within the sample set [37].

Genotype data

The procedure of DNA extraction from the Icelandic horse samples was described in the aforementioned genome-wide association studies [34, 35]. The 380 DNA samples were genotyped with the 670 K + Axiom Equine Genotyping Array. Quality control (QC) was performed using PLINK v1.9 software [38, 39]. For the ROH analysis, poorly genotyped data was removed based on criteria of missing genotypes per SNP (> 0.10) and missing SNPs per sample (> 0.10). No pruning for low minor allele frequency (MAF), deviation from Hardy-Weinberg equilibrium (HWE) or strong LD was done for the ROH analysis as recommended by Meyermans et al. [30]. Criteria for MAF (< 0.05) was however added when calculating the effective population size and observed and expected heterozygosity to be able to compare with similar studies. Only autosomal SNP markers were used for downstream analysis. The genotype data for the 280 Exmoor ponies was also derived from a 670 K Axiom Equine Genotyping Array. The same quality control criteria were used for the genotype data for the Exmoor ponies as for the Icelandic horse data described above. SNP positions were according to genomic coordinates in EquCab3.0 reference genome in both data sets.

After QC including pruning for MAF, the number of SNPs to be used for heterozygosity and Ne analyses was 360,755 and 322,209 for the Icelandic horses and Exmoor ponies, respectively. All the samples for the Icelandic horses passed QC, but six samples from the Exmoor pony group were discarded due to missing genotype data, leaving data for 274 Exmoor ponies for further analyses.

For the ROH analysis, QC was conducted without MAF pruning, resulting in 550,405 shared SNPs for downstream analysis using a combined dataset with information from both breeds.

Principal component analysis (PCA) was performed using the SNPRelate package [40] in R (version 4.3.1) [41] as a QC measure to identify outliers or sample mix-ups in the data as well as to visualize the genetic relationships and clustering patterns in the two breeds. The PCA plot, highlighting the distinct genetic signatures of the two breeds, is presented in Additional file 1: Fig. S1.

Pedigree analysis

Pedigree data for the Icelandic horse was obtained from the international Worldfengur database [8]. The pedigree data contained information about individuals born from 1860 to 2023, but the earliest records only included a small proportion of the population at that time. The quality of the pedigree data was estimated by calculating the pedigree completeness using the optiSel package [42] in R (version 4.3.1) [41]. The optiSel package was also used to estimate FPED and Ne based on the pedigree data for the 380 Icelandic horses in this study. This package estimates Ne from the mean rate of increase in coancestry. Velie et al. [37] reported that the pedigree data for the Exmoor ponies was complete for four generations. This pedigree data, of lower depth than that for the Icelandic horses, was not available for pedigree analysis in the present study and therefore we focused on genomic analysis for the Exmoor ponies.

Heterozygosity and effective population size trajectory

Trends in recent Ne trajectories were determined for both breeds using the SNeP v1.1 software [43]. Only Icelandic horses born between 2006 and 2016 (342 horses), and Exmoor ponies born between 1999 and 2009 (148 ponies) were used for the Ne analysis, covering approximately one generation interval. The minimum and maximum distance between pairs of SNPs was set to 0.05 Mb and 40 Mb, respectively, and the alpha value for the formula by Corbin et al. [44] used by the software to estimate Ne from LD was set to 2.2. The recombination rate was furthermore set to 1.24 × 10− 8, and the Sved & Feldman approximation [45] was used as a recombination rate modifier. The default value of 0.05 was used as minimum MAF.

Observed (HO) and expected heterozygosity (HE) was estimated for all the Icelandic horses and all the Exmoor ponies using the --het command in PLINK v1.9 [38, 39].

Runs of homozygosity and genomic inbreeding

The detectRUNS package [46] in R (version 4.3.1) [41] was used for analysing ROH with a sliding windows approach. The scanning window size was set equal to 10 SNP loci, and the maximum number of heterozygous or missing SNP in the sliding window was set equal to 0. The ROH parameter settings were optimised following recommendations in Meyermans et al. [30]. The final definition of the settings was as follows: (i) maximum distance between consecutive SNPs equal to 100 kb, (ii) minimum SNP density equal to 0.05 SNP/kb, (iii) minimum number of SNP in a run equal to 10 and (iv) minimum length of a run equal to 100 kb. One missing and one heterozygous SNP was allowed per run. The settings allowed ROH detection for 99.4% of the autosomal genome, indicating high validity of the analysis [30]. The minimum length of a run did not affect the genome coverage. Therefore, it was chosen based on the correlations between the FROH and FPED values for the Icelandic horses, which was highest (r = 0.57, p < 2.2 × 10− 16) when the minimum ROH length was set equal to 100 kb. The identified ROH were divided into five length classes (0.1 < ROH ≤ 1 Mb; 1 < ROH ≤ 2 Mb; 2 < ROH ≤ 4 Mb; 4 < ROH ≤ 8 Mb; and ROH > 8 Mb).

The FROH was calculated by summing each individual’s total length of ROH and dividing it by the autosomal genome length [24], which was set equal to 2281 Mb, based on the genome length covered by SNPs. FROH was calculated for each chromosome, length class, and as an average coefficient across the genome for each breed. Furthermore, to facilitate comparison with results from other similar studies, we also calculated the FROH values for both breeds when the minimum length of a run was set equal to 500 kb instead of 100 kb.

Signatures of selection and gene ontology

ROH islands that were shared by over 70% of the horses in each breed were determined as signatures of selection for that breed. A threshold of 70%, which is conservative compared to values found in the literature [12, 17, 18, 21,22,23, 25, 26, 47,48,49], was used to avoid false positive signatures of selection caused by population history events, such as genetic bottlenecks. The EquCab3.0 genomic coordinates of these regions were used to retrieve candidate gene lists from the genome browser Ensembl (release 110, July 2023) [50]. The candidate gene lists were subjected to a gene ontology (GO) analysis using PANTHER v18.0 (released Aug 2023) [51] to determine significantly enriched biological processes and molecular functions positively selected for in the breeds. Further functional annotation of possible candidate genes was performed using the GeneCards database (version 5.18, Oct 2023) [52, 53]. In addition, the Horse QTLdb (release 51, Aug 2023) [54 was used to identify any overlap with previously identified quantitative trait loci (QTL) in horses.

Results

Pedigree analysis

The evaluation of pedigree quality in the Icelandic horse dataset was based on the average number of discrete generation equivalents, resulting in a value of 8.21 (range: 6.15 to 10.1). This value signifies good pedigree completeness. The average FPED for the 380 Icelandic horses in this study was estimated to be 0.03. The comparison between FROH and FPED revealed a linear relationship (r = 0.57, p < 2.2 × 10− 16) (Fig. 1).

Fig. 1
figure 1

Correlation between FROHand FPED.

Pearson correlation between FROH and FPED in the Icelandic horse, with a 95% confidence interval (grey area)

Full size image

Heterozygosity and effective population size trajectory

The overall mean HO and HE in the Icelandic horse were equally estimated to be 0.34. In the studied data, HO values ranged from 0.30 to 0.38. Similarly, the HE estimated for the Exmoor pony was 0.34 but the mean HO was 0.33 and ranged between 0.19 and 0.41.

Based on genomic information, the Ne for the last generation of Icelandic horses was estimated to be approximately 125 individuals (Fig. 2). The trend exhibited an overall decline for the past 60 generations, with more pronounced decreases observed around 18–23 and 7–8 generations ago. However, in the most recent 3–4 generations, the Ne trend levelled off, fluctuating within the range of 123 to 127 individuals. In contrast, the genomic Ne estimate for the Exmoor pony was 42 individuals in the last generation. Furthermore, the trend observed for the Exmoor pony over the last 60 generations indicates a consistent, albeit gradual, decrease in Ne.

Fig. 2
figure 2

Trends in effective population size

Changes in effective population size of the Icelandic horse breed and the Exmoor pony over the last 60 generations based on genomic information

Full size image

Runs of homozygosity and genomic inbreeding

A total of 573,746 and 548,302 ROH were identified for the Icelandic horse and the Exmoor pony, respectively (Table 1). In both cases, the majority of the identified ROH (≥ 96%) was categorised in the shortest length class (0.1 to ≤ 1 Mb) with an average ROH length of 0.24 Mb in the Icelandic horse and 0.26 Mb in the Exmoor pony. The average occurrence of the short ROH was 1455 per individual in the Icelandic horse data and 1921 per individual in the Exmoor pony data. Only 125 Icelandic horses, out of the 380, carried ROH islands categorised in the longest length class (> 8.0 Mb), and on average they carried 2 such ROH islands with a mean length of 10.8 Mb. On the other hand, only one ROH in a single Exmoor pony was identified to belong to the longest length class.

Table 1 Descriptive variables from the ROH analysis of the Icelandic horse genome and the Exmoor pony genome
Full size table

ROH quantity, distribution, and average length were estimated per chromosome in both breeds (Fig. 3). The analysis revealed the highest number of ROH in both breeds on ECA1 (NROH = 47,793 in the Icelandic horse, NROH = 43,203 in the Exmoor pony), and the lowest on ECA31 (NROH = 6212 in the Icelandic horse, NROH = 6227 in the Exmoor pony). For Icelandic horses, ECA23 had the longest average ROH (LROH = 0.33 Mb), while the shortest (LROH = 0.25 Mb) were found on ECA12. In Exmoor ponies, the longest average ROH was on ECA22 (LROH = 0.36 Mb), and the shortest on ECA20 (LROH = 0.25 Mb).

Fig. 3
figure 3

Distribution and average length of ROH across chromosomes

(a) Distribution and (b) average length of ROH in Mb detected across the autosomal genome in the Icelandic horse and the Exmoor pony

Full size image

The estimated mean FROH was relatively high in both breeds, with a total of 0.20 in the Icelandic horse and 0.27 for the Exmoor pony (Table 2) when including ROH lengths from 100 kb and higher. The individual FROH ranged from 0.07 to 0.30 for the Icelandic horses, and from 0.01 to 0.55 for the Exmoor ponies. The distribution of average FROH values across the genome in both breeds is shown in a violin plot in Fig. 4. FROH estimations for the different ROH length classes revealed that most of the inbreeding could be traced back to the high amount of ROH identified in the shortest length class (0.1 to ≤ 1 Mb) in both breeds. The length classes comprising longer ROH (> 4 to ≤ 8 Mb, and > 8 Mb) accounted for low amount of inbreeding in the Icelandic horse, with FROH values of 0.01 for each of these two ROH length classes, respectively. In the Exmoor pony, close to zero inbreeding was estimated based only on longer ROH (> 4 Mb).

Fig. 4
figure 4

Violin plot showing distribution of genome-wide FROH

Distribution of average FROH across the genome for the Icelandic horse (to the left) and the Exmoor pony (to the right) represented with a violin plot including a box plot indicating the median, first and third quartile (Q1 and Q3) and the outliers

Full size image
Table 2 Results for each breed’s mean FROH across the genome and the mean FROH across the five length classes
Full size table

The analysis across chromosomes for the Icelandic horse revealed the highest mean FROH, including all ROHs (≥ 0.1 Mb), on ECA23 (FROH = 0.27 ± 0.09, max = 0.74, min = 0.10) and the lowest mean FROH on ECA12 (FROH = 0.15 ± 0.07, max = 0.56, min = 0.02) and ECA20 (FROH = 0.15 ± 0.07, max = 0.58, min = 0.03). For the Exmoor pony, the highest mean FROH was identified on ECA22 (FROH = 0.34 ± 0.17, max = 0.97, min = 0.01) and ECA23 (FROH = 0.35 ± 0.17, max = 0.92, min = 0.00) and the lowest mean FROH on ECA12 (FROH = 0.22 ± 0.13, max = 0.67, min = 0.00) and ECA20 (FROH = 0.21 ± 0.13, max = 0.78, min = 0.00). A violin plot of mean genomic inbreeding across chromosomes within each breed is shown in Additional file 2: Fig. S2.

When the minimum length of ROH was set equal to 500 kb instead of 100 kb, and thus not including the shortest ROH (0.1–0.5 Mb), the average FROH for the Icelandic horse was 0.08 and it was 0.12 for the Exmoor pony. Details of FROH estimates within different length classes from this analysis are shown in Additional file 3: Table S1.

Signatures of selection and gene ontology

A total of 15 chromosomes (ECA1, ECA3-5, ECA7-9, ECA11-12, ECA17-20, ECA23, and ECA29) contained ROH islands that were shared by more than 70% of the individuals in the Icelandic horse sample (Fig. 5a), while a total of 23 chromosomes (ECA1-9, ECA11-12, ECA14-19, ECA21-24, and ECA30-31) contained ROH islands shared by more than 70% of the Exmoor ponies (Fig. 5b). The most prominent ROH island hot spot in the Icelandic horse, shared by over 90% of the individuals was located on ECA23 in the region where the DMRT3 gene is located. On the other hand, the most prominent ROH island hot spot in the Exmoor pony, also shared by over 90% of the sampled individuals, was on ECA22. A complete list of all identified ROH islands for both breeds is shown in Additional file 4: Table S2. The two breeds had overlapping ROH islands on six chromosomes (ECA1, ECA3, ECA8, ECA9, ECA17 and ECA19). The list of annotated genes within and in the vicinity of these shared ROH islands is shown in Additional file 5: Table S3.

Fig. 5
figure 5

Manhattan plot showing ROH islands across chromosomes

ROH islands across all autosomes in (a) the Icelandic horse and (b) the Exmoor pony. The x-axis represents the chromosome numbers, and the y-axis represents the proportion of animals sharing a ROH. The ROH islands exceeding the 70% threshold (red dotted line) were considered signatures of selection

Full size image

Overall, 37 annotated genes were located within the identified ROH islands in the Icelandic horse and 289 more in their ± 500 kb vicinity. In the Exmoor pony, 181 annotated genes were identified within the ROH islands, and 645 more in their ± 500 kb vicinity. Given the considerable number of ROH islands detected, we decided to concentrate on specific regions that could be associated with traits in either or both of the studied breeds in this article.

Table 3 presents the private ROH islands identified in the Icelandic horse that were linked to performance traits, and genes within those ROH. Similarly, Table 4 outlines the private ROH islands identified in the Exmoor ponies and associated genes that are related to coat colours, fertility, hypertension, and the immune system. Additionally, Table 5 enumerates the shared ROH islands between the two breeds suggested to be associated with the immune system, metabolism, coat type, and body size.

Table 3 List of selected private ROH islands in the Icelandic horses with annotated genes located within each ROH island and traits possibly associated with the genes or regions
Full size table
Table 4 List of selected private ROH islands in the Exmoor pony with annotated genes located within each ROH island and traits possibly associated with the genes or regions
Full size table
Table 5 List of selected shared ROH islands in the Icelandic horses (ICE) and the Exmoor pony (EXM) with annotated genes located within each ROH island and traits possibly associated with the genes or regions
Full size table

Discussion

Disentangling genomic adaptation from natural and artificial selection within a genome is a challenging task. Some approaches include population genetic analysis, functional genomics, examination of historical data on breeding practices, and genomic comparisons. The Icelandic horse and Exmoor pony share many similarities, being ancient breeds of relatively small body size adapted to harsh conditions, but they also exhibit significant differences due to stringent selection for gait performance in Icelandic horses and emphasis on coat colour and conservation of Exmoor ponies. The comparison of the two breeds, therefore, gives a valuable opportunity to distinguish between detected signatures of selection for performance, and signatures resulting from adaptations to harsh environment. Furthermore, this study is the first to analyse genetic diversity and ROHs in the Icelandic horse using high-density SNP-marker data.

A substantial number of ROH islands, considered indicative of selection signatures, were identified in both breeds, and presented in Additional file 4: Table S2. However, in this discussion we focus on selected regions linked to specific traits relevant to the studied breeds. Shared ROH islands between both breeds were identified, associated with metabolic processes, body size, and the immune system. Additionally, private ROH islands linked to performance in the Icelandic horse and ROH islands associated with coat colours, hypertension, and fertility in the Exmoor pony were identified.

Exploring genetic diversity

Our findings revealed similar heterozygosity estimates in both the Icelandic horse (mean HO = 0.34) and the Exmoor pony (mean HO = 0.33), which were also comparable to estimates reported in other breeds with closed populations. In a study by Cosgrove et al. [16], which estimated heterozygosity for various breeds in the development of the 670k genotyping array [55], reported average HO = 0.34 ± 0.02 for pony breeds, HO = 0.33 ± 0.01 for draft horse breeds, and HO = 0.32 ± 0.01 for the Icelandic horse.

Estimations of SNP based Ne of the Icelandic horse (125 individuals) further suggested an adequate genetic diversity within the breed, and a relative stability in Ne estimations over the last 3–4 generations. The equal estimations of HE and HO and the low number of longer ROH support the stability observed in the Ne trend in recent generations, and the absence of strong recent inbreeding. In contrast, the Ne estimate for the Exmoor pony (42 individuals) indicates severe loss of genetic diversity, posing a risk to the breed’s sustainability. This decline is likely linked to the grave bottleneck experienced during the Second World War when the pony population diminished to about 50 individuals by the war’s end [9]. While the bottleneck may not be explicitly evident in the Ne trend depicted in Fig. 2, the gradual, slow decrease suggests that conservation measures applied after the war may have been useful. However, there were a few Exmoor ponies with unexpectedly low FROH in the present study and the subpopulations that Velie et al. [37] identified in the Exmoor data, the relatively small sample size, and the lack of background information prevented us from drawing any definitive conclusions for the Ne trend.

Seen over a longer time span for the Icelandic horse, Ne estimations revealed a significant decrease at two historical events, resulting in a major decline in the population size of the Icelandic horse [7]. The first event occurred approximately 23 generations ago, aligning with the end of the 18th century considering a generation interval of roughly 10 years, coinciding with the Skaftáreldar volcanic eruption. This eruption had detrimental consequences for both humans and livestock in Iceland, and is said to have reduced the number of horses with 75% [56]. The second event took place around 8 generations ago, during the industrial revolution in Iceland. During this period, the role of the Icelandic horse as a working horse was superseded by machines, leading to a shift in the breeding goal towards breeding riding horses [4, 57].

While the Ne estimates for the Icelandic horse exceed the generally recommended minimum size for sustaining genetic diversity in selectively bred populations, previous studies have shown a decline in pedigree-based Ne over generations. For instance, in 1989, the pedigree-based Ne was reported as 365 individuals [58], declining to 210 animals by 2000 [59], and currently estimated at around 100 individuals [27]. These figures differ somewhat with the Ne estimates from genomic data presented in Fig. 2 of this study. The pedigree completeness for the Icelandic horse is high, but pedigree measures are limited to probability estimates based on documented ancestry. In contrast, genomic data provides a more detailed and comprehensive view of the population’s genetic structure and history but is based on a genotyped subset of the population. Therefore, some differences in pedigree-based and genomic estimates can be expected. However, it is reassuring that the differences were not substantial.

The average FPED estimated in the present study closely aligns with recent calculations for all horses born in Iceland between 2011 and 2020 [27]. This suggests that our sample predominantly consisting of preselected Icelandic breeding horses likely to contribute to future generations fairly accurately represents the latest generation of Icelandic horses. A continued monitoring of relatedness and genetic contribution of breeding animals, and resulting inbreeding trends, is important to ensure a sustainable breeding program. This is especially important within the Icelandic horse population, because a large part of the population is geographically isolated in Iceland where importation of genetic material is prohibited according to the Animal Importation Act [60].

Interpreting genomic inbreeding

The quantity of detected ROH islands in this study was heavily influenced by the parameter settings, notably the minimum ROH length. Many equine studies using 670k SNP data set a minimum ROH length of 500 kb [17, 19, 21, 22, 26], excluding the shortest ROHs (< 0.5 Mb). This exclusion further impacts FROH estimation, which is derived from the total genome length covered by ROHs. In our study, a 100 kb minimum ROH length resulted in the highest correlation between estimated FROH and FPED values for the Icelandic horse. Using this setting, the mean FROH was higher (FROH = 0.20 for the Icelandic horse and FROH = 0.27 for the Exmoor pony) compared to previous studies. Previous reports for the Exmoor pony ranged from 0.17 to 0.25 [12, 18], while the disparity in FROH values for the Icelandic horse was more pronounced, with previous estimates ranging from 0.03 to 0.13 [16, 18, 30]. Excluding the shortest ROHs (< 0.5 Mb), resulted in values closer to those previously reported. It can be argued that the FROH estimate based on the longer ROHs, reflecting more recent inbreeding, is more relevant for risk assessment of the current breeding practises [61, 62].

The high average FROH observed in the Icelandic horse when including the shortest ROHs, may be attributed to the breed’s assumedly limited initial genetic pool, potential drift, and genetic purging during its adaptation process. The relatively small number of ROHs longer than 4 Mb suggests no evidence of recent excessive inbreeding. However, when compared with the near absence of long ROHs in the Exmoor pony, there is suggestive evidence of a stronger recent selection in the Icelandic horse. Additionally, there are indications of an increased contribution of a limited number of breeding animals in the Icelandic horse to the modern gene pool [27], emphasizing the importance of closely monitoring inbreeding and genetic diversity in the breed.

ROH island cold spots of different origin

ROH island cold spots were identified on ECA12 and ECA20 in both breeds, where the shortest average ROHs and the lowest mean FROH were found. The cold spot on ECA20 may be attributed to the major histocompatibility complex (MHC) covering a substantial portion of the chromosome [63, 64]. The MHC is a highly variable region associated with the immune system and benefits from heterozygosity [65], with the Icelandic horse, for instance, showing high MHC heterozygosity [66]. Furthermore, a possible cause of the cold spot identified on ECA12 is the higher percentage of the chromosome covered by copy number variation (CNV) gains and losses compared to other equine chromosomes [67,68,69]. CNV increases genetic diversity by varying the number of copies of genomic regions [70], indicating higher heterozygosity in the region on ECA12.

Performance-linked ROH islands on ECA23 and ECA17 in the Icelandic horse

The most prominent ROH island hot spot in the Icelandic horse was located on ECA23; a region harbouring genes such as the DMRT3 and DOCK8, both known to be causative or highly associated with gaits and performance in many horse breeds [71,72,73,74,75,76,77,78]. A single mutation [DMRT3:Ser301STOP marker at nucleotide position 22,999,655 on ECA23] in the DMRT3 gene, also referred to as the ‘Gait keeper’ mutation, alters the pattern of locomotion and has a predominant effect on gaiting ability in Icelandic horses [71, 72]. The identified ROH harbouring the DMRT3 gene was the longest ROH (589 kb) identified in this study that was shared by over 70% of the Icelandic horses, indicating recent selection for this region.

This ROH also harbours the DMRT1, DMRT2 and KANK1 genes and overlaps the ‘Gait keeper’ haplotype previously identified [71, 79]. Furthermore, this region overlaps a previously identified selection signature for the Icelandic horse in a study by Petersen et al. [31]. The DOCK8 gene, located in another ROH (106 kb) on ECA23, has been shown to be associated with harness racing success in Nordic trotters [73]. Furthermore, in a small sample set of Icelandic horses, the DOCK8 gene was found to be associated with pace racing success and to potentially segregate between elite pace racers and other horses [80]. Previous studies hypothesized overlapping or common gene effects of the DMRT1-3 genes and the DOCK8 gene [73, 81, 82].

Even though the ‘Gait keeper’ mutation has been shown to be a causative factor for gaiting ability, it is highly unlikely that it is the single cause as shown by multiple studies [35, 71, 72, 74, 76,77,78, 83]. It is therefore possible that the DOCK8 gene contributes to the performance of gaits, alongside the DMRT3 gene. The PGM5 gene, located in the third ROH (112 kb) on ECA23, has no known association with performance in horses. It is predicted to enable metal ion binding activity and phosphoglucomutase activity, and to be associated with myofibril assembly and striated muscle tissue development in zebrafish [84], and may thus be a candidate to study further for performance in horses.

A relatively long ROH island (233 kb) was detected on ECA17 for the Icelandic horses, harbouring the genes EDNRB and SLAIN1. The EDNRB gene harbours the ‘Overo allele’, which has been shown to be the causative factor for the Overo coat colour in horses and the lethal white foal syndrome (LWFS) in homozygous form [85,86,87]. Since there are no reports of either the Overo coat colour or the LWFS in the Icelandic horse breed, the apparent selection intensity for this region is likely associated with another function of the gene EDNRB, that appears to have pleiotropic effects. The EDNRB gene is a part of the endothelin gene family, which plays a crucial role in regulating blood vessel tone and blood pressure [88, 89]. The EDNRB interacts with its family members, such as the EDN3 gene [90,91,92] suggested to be associated with blood supply regulation in high-performing racing horses [93, 94]. Icelandic horses are trained for high intensity exercises [95, 96], indicating the importance of a robust regulatory system for the distribution of blood to the tissues. This ROH island may, therefore, be a product of selection for performance. The SLAINI gene has furthermore been associated with the developing nervous system in mouse embryos [97], indicating a possible importance for performance.

Another possible performance related ROH island was detected on ECA1, where the NEURL1 and SH3PXD2A genes are located. One of the functions of the NEURL1 gene is hippocampal-dependent synaptic plasticity, which affects learning and memory processes [98, 99]. This region could, therefore, be important for horses trained for performance. These ROH islands on ECA23, ECA17 and ECA1 were not identified in the Exmoor pony genome in this study, further underlining the possible association with performance.

ROH islands distinguished by coat colour genes in the Exmoor pony

The Exmoor pony is renowned for its distinctive bay coat colour and mealy markings. A ROH island hot spot was identified on ECA22, coinciding with the location of the ASIP gene which is responsible for the bay coat colour in horses [100]. Additionally, a prominent signature on ECA8, harbouring the TBX3 and TBX5 genes, was observed for this breed. While TBX3 controls dun coat colour [101], the rarity of dun-coloured Exmoor ponies suggests that the signature likely reflects the high prevalence of the non-dun alleles in the gene.

The OCA2 gene was identified in a ROH island on ECA1 for Exmoor ponies and is known to be one of the components of the mammalian pigmentary system [102,103,104]. The gene is a major determinant of brown and/or blue eye colour [103,104,105] and is hypothesized to be a key control point at which ethnic skin colour variation in humans is determined [106]. Efforts have been made to link this gene to horse colour phenotypes [107, 108] without success so far. The OCA2 gene consistently emerges as a selection signature in the Exmoor pony genome [17], suggesting its potential association with some of their characteristics, such as the mealy markings. The mealy phenotype has previously been linked to the EDN3 gene [93], which was also identified in this study, located near another ROH island on ECA22. Consequently, we recommend further exploration of these two candidate genes to ascertain their potential association with the mealy phenotype.

No ROH islands harbouring candidate genes for horse colour phenotypes were identified in the Icelandic horse genome. This absence may be attributed to the breeding goal for the Icelandic horse [109], which has consistently aimed at preserving a diverse range of coat colours, presumably leading to higher variability within the colour loci.

Signs of adaptation to limited feed supply

One of the most prominent ROH islands shared by the Icelandic horse and the Exmoor pony, harboured the FOXO1 gene on ECA17. The FOXO1 gene has been associated with insulin resistance [110,111,112] which is one of the key components of the equine metabolic syndrome (EMS) [113]. EMS is generally observed in breeds categorized as “easily fed,” which typically require a lower nutritional intake to maintain body weight. These breeds, including the Icelandic horse and Exmoor pony, have historical backgrounds marked by poor feed availability and periods of starvation. The hypothesis suggests that positive selection for this genomic region has historically contributed to the survival of these breeds in harsh winter conditions but may render them less adaptive to lush pastures and high-energy diets, and in some cases, low workload. Low insulin sensitivity, or even insulin resistance, has been reported in both breeds [114, 115].

Another key aspect of EMS involves a susceptibility to laminitis [113], which has been shown to be associated with hypertension in horses [115]. Moreover, hypertension arises from dysfunction in vascular endothelial cells in humans with type 2 diabetes [116], a syndrome considered closely related with EMS [117]. Additionally, the vascular endothelium plays a crucial role in preventing platelet activation and the adhesion of leukocytes to the vascular wall [115]. Within the Exmoor pony genome, a substantial homozygote region on ECA22 was identified, harbouring three distinct ROH islands. In the first ROH, the genes BMP7 and SPO11 were identified; the second contained the NPEPL1 and GNAS genes, while the third encompassed the NELFCD, CTSZ, TUBB1, and PRELID3B genes. Notably, all genes in the third ROH are associated with the regulation of platelet properties [52, 53, 118, 119]. Moreover, research has linked the BMP7 gene to diabetes and vascular calcification in humans [120], while the SPO11 gene has been linked to endothelial dysfunction resulting from exercise-induced DNA damage in horses [94]. Furthermore, mutations in the GNAS gene have been established as causative for McCune-Albright syndrome in humans, a condition known to involve endocrinologic anomalies such as Cushing syndrome [121]. Equine Cushing’s disease is recognized in many horse breeds and frequently leads to the development of laminitis [117]. At last, the aforementioned EDN3 gene, which is a part of the endothelin gene family, is located in a close proximity (> 165 kb) to the ROH islands.

The strong evidence of genes associated both directly and indirectly with vascular endothelin regulation in the specified ECA22 region suggests it could be a signature for positive selection, representing an adaptive trait in Exmoor ponies potentially related to varying feed supply. This study did not identify evidence of positive selection for the same region on ECA22 in the Icelandic horse, however.

Hot spot on ECA11 potentially linked to harsh climate adaptation

ROH islands were identified on ECA11 in both the Icelandic horse (25,277,282 − 29,237,132) and Exmoor pony (30,266,841 − 31,495,910) within a region that appears to be partly shared among various pony and draft horse breeds [17, 18, 21,22,23, 31, 47,48,49]. This region has been shown to have a low recombination rate in horses [122]. It ranges from approximately position 23 Mb to 32 Mb and has predominantly been associated with phenotypes such as a small to medium height at withers, and a compact, muscular body and robust bone structure, as observed in pony and draft horse breeds [123,124,125], and has also been suggested to be involved in hair and coat density and quality [48, 49, 126].

Whereas phenotypes such as limited height at withers and dense winter coat apply to the Icelandic horse and Exmoor pony, the genes in this wider genomic region previously suggested to be of importance for such traits were not within the ROH islands identified on ECA11 in the present study. However, we cannot exclude that selection has targeted other nearby genes, given the overall high gene density in the region, or that regulatory functions have been selected for. The low recombination rate in the region [122] suggests strong linkage and perhaps participation of many genes in similar processes.

Adapted antibacterial defence

A ROH island identified on ECA9 (73,072,557 − 73,286,648) shared by both horse breeds may play a crucial role in the immune system. Within this region lies the CYRIB gene, which has been shown to be associated with protection against Salmonella bacterial infections in humans and contribute to restricting infections mediated by Mycobacterium tuberculosis and Listeria monocytogenes [127].

Functional annotation analyses of genes found within a ROH island on ECA15 only in the Exmoor pony (10,154,743 − 10,399,931) revealed an enrichment in GO terms related to the “defence response to Gram-positive bacterium”. Gram-positive bacteria include genera like Staphylococcus, Streptococcus, Clostridium, and Listeria, all known to cause diseases of varying severity in horses [128,129,130,131]. The genes identified within the ROH include REV1, EIF5B, TXNDC9, LYG1, and LYG2. The two last ones, LYG1 and LYG2, have been reported to have a significant role in innate immunity in mammals [132]. As far as our knowledge extends, this specific region has not been recognized as a selection signature in other horse breeds, while the LYG1 and LYG2 genes have been identified as candidate genes for selection in sheep [133].

Male fertility related genomic regions in the exmoor pony

Three ROH islands detected in the Exmoor pony genome harbour genes related to male fertility traits. First, a ROH island was detected on ECA11, positioned at 33,050,441 to 33,198,699, containing the TEX14 gene. Second, a ROH island was identified on ECA22 ranging from 44,748,478 to 44,900,296, harbouring the SPO11 gene. At last, a ROH detected on ECA24 (16,374,545 − 16,425,073) harboured genes that, by a functional annotation analysis, revealed an enrichment in the GO term related to “male gonad development”.

The TEX14 gene codes for a testis-specific protein and serves as a crucial element in the intercellular bridges of both male and female embryos. Adult male mice lacking TEX14 mRNA are unable to reproduce (sterile), while females with the same genetic condition maintain their fertility [134, 135]. TEX14 has further been suggested to have been targeted by selection for fertility in German warmblood horses [13] and was located within a ROH island detected in the Noriker horse genome [23]. The SPO11 gene codes for an evolutionarily conserved topoisomerase-like protein that, in mammals, is functionally expressed in gonads during meiosis. It has been shown to be associated with male infertility in mice, humans, and cattle [136,137,138,139,140]. The ROH island on ECA24 contained the MED6 gene and three novel genes (ENSECAG00000031483, ENSECAG00000036606, ENSECAG00000041394). ENSECAG00000041394 is an orthologue of the ADAM20 mouse gene. The ADAM metallopeptidase domain 20 (ADAM20) gene is specifically expressed in testis and has been associated with male infertility in humans and mice [141,142,143,144]. The presence of these ROH islands associated with male fertility implies that this trait may have undergone positive selection in the Exmoor pony breed, as a survival trait in semi-feral conditions. These ROHs were not detected in the Icelandic horse in this study.

Conclusions

This study provides insights into the genetic diversity and genomic ROH patterns in the Icelandic horse and Exmoor pony. Our assessments indicate that the genetic diversity in the Icelandic horse is on an acceptable level for a closed population undergoing artificial selection. Nevertheless, it is advisable to maintain ongoing monitoring to guarantee the preservation of genetic diversity and to support sustainable breeding practices for the Icelandic horse. In contrast, our results for the Exmoor pony indicates a critical state of genetic diversity. However, further research accounting for the population structure of the breed is needed to validate our findings.

The FROH estimates were significantly affected by the parameters employed in the ROH analysis, emphasizing the importance of considering these settings when comparing values across different studies. In our study, the high occurrence of short ROHs led us to attribute a larger extent of the identified inbreeding in both breeds to historical events like the breed’s origin, bottlenecks, and adaptation, rather than recent and stringent selection practices.

Several ROH islands associated with performance were identified in the Icelandic horse, effectively distinguishing the breed from the Exmoor pony. The most prominent one on ECA23 featured the longest average ROHs and the highest mean FROH across all chromosomes, suggesting the most recent and stringent selection pressure. The shared ROH islands observed in both breeds were linked to traits associated with adapting to challenging environments with limited food resources, as well as to immune system function. Conversely, distinct ROH regions specific to Exmoor ponies were associated with their exterior characteristics such as coat colour, along with traits related to immune response and fertility.

In conclusion, this study provides knowledge contributing to preserving genetic diversity and population health in these two equine populations. Furthermore, the obtained results provide important insight into genomic regions shared by the two breeds, which are likely associated with adaptive traits shaped by natural selection. Genomic regions related to performance were identified only in the Icelandic horse, likely reflecting the artificial selection for gaits and performance that has occurred over the past few decades.

Data availability

This study did not generate new data; all data used were pre-existing. The Icelandic horse genotypes analysed during the study have been deposited in the European Variation Archive (EVA) [145] at EMBL-EBI under accession number PRJEB74212 (https://www.ebi.ac.uk/eva/?eva-study=PRJEB74212). The Exmoor pony genotypes are available via Figshare (DOI: https://doi.org/10.6084/m9.figshare.3145759).

Abbreviations

BLUP:

Best linear unbiased prediction

CNV:

Copy number variation

DNA:

Deoxyribonucleic acid

ECA:

Equus caballus chromosome

EMS:

Equine metabolic syndrome

GO:

Gene ontology

HWE:

Hardy-Weinberg equilibrium

LD:

Linkage disequilibrium

LWFS:

Lethal white foal syndrome

MAF:

Minor allele frequency

MHC:

Major histocompatibility complex

mRNA:

Messenger RNA

PCA:

Principal component analysis

QC:

Quality control

QTL:

Quantitative trait loci

RNA:

Ribonucleic acid

ROH:

Runs of homozygosity

SD:

Standard deviation

SNP:

Single nucleotide polymorphism

References

  1. Eding H, Bennewitz J. Measuring genetic diversity in farm animals. In: Oldenbroek K, editor. Utilisation and conservation of farm animal genetic resources. 1st ed. Wageningen: Wageningen Academic; 2007. pp. 103–30.

    Chapter  Google Scholar 

  2. Wolliams J, Berg P, Mäki-Tanila A, Meuwissen T, Fimland E. Sustainable management of animal genetic resources. Ås (NO): Nordisk Genbank Husdyr; 2005.

    Google Scholar 

  3. Adalsteinsson S. Origin and conservation of farm animal populations in Iceland. Z Tierz Züchtungsbio. 1981;98:258–64.

    Article  Google Scholar 

  4. Arnórsson K. Ræktunin [The breeding]. In: Björnsson GB, Sveinsson HJ, editors. Íslenski hesturinn [The Icelandic horse]. 1st ed. Reykjavík: Mál og menning; 2006. pp. 202–47.

    Google Scholar 

  5. Árnason T. Genetic studies on conformation and performance of Icelandic toelter horses [dissertation]. Uppsala: Reklam & katalogtryck; 1983 [cited 2024 February 11].

  6. Árnason T. Genetic studies on conformation and performance of Icelandic toelter horses: IV. Best linear unbiased prediction of ten correlated traits by use of an animal model. Acta Agr Scand. 1984;34:450–62.

    Article  Google Scholar 

  7. Hreiðarsdóttir GE, Árnason Þ, Svansson V, Hallsson JH. Analysis of the history and population structure of the Icelandic horse using pedigree data and DNA analyses. Icel Agric Sci. 2014;27:63–79.

    Google Scholar 

  8. Worldfengur - The studbook of origin for the Icelandic horse. Farmers Association of Iceland and FEIF International Federation of Icelandic Horse, Reykjavik. 2001. https://www.worldfengur.com. Accessed 11 February 2024.

  9. The Exmoor Pony Society. https://exmoorponysociety.org.uk. Accessed 30 May 2024.

  10. Curik I, Ferenčaković M, Sölkner J. Inbreeding and runs of homozygosity: a possible solution to an old problem. Livest Sci. 2014;166:26–34.

    Article  Google Scholar 

  11. Peripolli E, Munari DP, Silva MVGB, Lima ALF, Irgang R, Baldi F. Runs of homozygosity: current knowledge and applications in livestock. Anim Genet. 2017;48:255–71.

    Article  CAS  PubMed  Google Scholar 

  12. Ablondi M, Viklund Å, Lindgren G, Eriksson S, Mikko S. Signatures of selection in the genome of Swedish warmblood horses selected for sport performance. BMC Genomics. 2019;20:717.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Nolte W, Thaller G, Kuehn C. Selection signatures in four German warmblood horse breeds: tracing breeding history in the modern sport horse. PLoS ONE. 2019;14:e0215913.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Hill EW, Stoffel MA, McGivney BA, MacHugh DE, Pemberton JM. Inbreeding depression and the probability of racing in the Thoroughbred horse. P Roy Soc B-Biol Sci. 2022;289:20220487.

    CAS  Google Scholar 

  15. Santos WB, Schettini GP, Maiorano AM, Bussiman FO, Balieiro JCC, Ferraz GC, et al. Genome-wide scans for signatures of selection in Mangalarga Marchador horses using high-throughput SNP genotyping. BMC Genomics. 2021;22:737.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Cosgrove EJ, Sadeghi R, Schlamp F, Holl HM, Moradi-Shahrbabak M, Miraei-Ashtiani SR, et al. Genome diversity and the origin of the arabian horse. Sci Rep. 2020;10:9702.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Grilz-Seger G, Neuditschko M, Ricard A, Velie B, Lindgren G, Mesarič M, et al. Genome-wide homozygosity patterns and evidence for selection in a set of European and Near Eastern horse breeds. Genes-Basel. 2019;10:491.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Colpitts J, McLoughlin PD, Poissant J. Runs of homozygosity in Sable Island feral horses reveal the genomic consequences of inbreeding and divergence from domestic breeds. BMC Genomics. 2022;23:501.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Sadeghi R, Moradi-Shahrbabak M, Miraei Ashtiani SR, Schlamp F, Cosgrove EJ, Antczak DF. Genetic diversity of persian arabian horses and their relationship to other native Iranian horse breeds. J Hered. 2018;110:173–82.

    Article  Google Scholar 

  20. Laseca N, Molina A, Ramón M, Valera M, Azcona F, Encina A et al. Fine-scale analysis of runs of homozygosity islands affecting fertility in mares. Front Vet Sci. 2022;9.

  21. Ablondi M, Dadousis C, Vasini M, Eriksson S, Mikko S, Sabbioni A. Genetic diversity and signatures of selection in a native Italian horse breed based on SNP data. Animals. 2020;10:1005.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Grilz-Seger G, Druml T, Neuditschko M, Dobretsberger M, Horna M, Brem G. High-resolution population structure and runs of homozygosity reveal the genetic architecture of complex traits in the Lipizzan horse. BMC Genomics. 2019;20:174.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Grilz-Seger G, Druml T, Neuditschko M, Mesarič M, Cotman M, Brem G. Analysis of ROH patterns in the Noriker horse breed reveals signatures of selection for coat color and body size. Anim Genet. 2019;50:334–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. McQuillan R, Leutenegger A-L, Abdel-Rahman R, Franklin CS, Pericic M, Barac-Lauc L, et al. Runs of homozygosity in European populations. Am J Hum Genet. 2008;83:359–72.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Bizarria dos Santos W, Pimenta Schettini G, Fonseca MG, Pereira GL, Loyola Chardulo LA, Rodrigues Machado Neto O, et al. Fine-scale estimation of inbreeding rates, runs of homozygosity and genome-wide heterozygosity levels in the Mangalarga Marchador horse breed. J Anim Breed Genet. 2021;138:161–73.

    Article  CAS  PubMed  Google Scholar 

  26. Mousavi SF, Razmkabir M, Rostamzadeh J, Seyedabadi H-R, Naboulsi R, Petersen JL, et al. Genetic diversity and signatures of selection in four indigenous horse breeds of Iran. Heredity. 2023;131:96–108.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Árnadóttir E. Erfðafjölbreytileiki íslenska hrossastofnsins (Genetic diversity of the Icelandic horse population). BSc thesis, Agricultural University of Iceland. 2022.

  28. McCue ME, Bannasch DL, Petersen JL, Gurr J, Bailey E, Binns MM, et al. A high density SNP array for the domestic horse and extant perissodactyla: utility for association mapping, genetic diversity, and phylogeny studies. PLOS Genet. 2012;8:e1002451.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Petersen JL, Mickelson JR, Cothran EG, Andersson LS, Axelsson J, Bailey E, et al. Genetic diversity in the modern horse illustrated from genome-wide SNP data. PLoS ONE. 2013;8:e54997.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Meyermans R, Gorssen W, Buys N, Janssens S. How to study runs of homozygosity using PLINK? A guide for analyzing medium density SNP data in livestock and pet species. BMC Genomics. 2020;21:94.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Petersen JL, Mickelson JR, Rendahl AK, Valberg SJ, Andersson LS, Axelsson J, et al. Genome-wide analysis reveals selection for important traits in domestic horse breeds. PLOS Genet. 2013;9:e1003211.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Gorssen W, Meyermans R, Janssens S, Buys N. A publicly available repository of ROH islands reveals signatures of selection in different livestock and pet species. Genet Sel Evol. 2021;53:2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Shrestha M, Solé M, Ducro BJ, Sundquist M, Thomas R, Schurink A, et al. Genome-wide association study for insect bite hypersensitivity susceptibility in horses revealed novel associated loci on chromosome 1. J Anim Breed Genet. 2020;137:223–33.

    Article  CAS  PubMed  Google Scholar 

  34. Rosengren MK, Sigurðardóttir H, Eriksson S, Naboulsi R, Jouni A, Novoa-Bravo M, et al. A QTL for conformation of back and croup influences lateral gait quality in Icelandic horses. BMC Genomics. 2021;22:267.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Sigurðardóttir H, Boije H, Albertsdóttir E, Kristjansson T, Rhodin M, Lindgren G, et al. The genetics of gaits in Icelandic horses goes beyond DMRT3, with RELN and STAU2 identified as two new candidate genes. Genet Sel Evol. 2023;55:89.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Albertsdóttir E, Eriksson S, Sigurdsson Á, Árnason T. Genetic analysis of ‘breeding field test status’ in Icelandic horses. J Anim Breed Genet. 2011;128:124–32.

    Article  PubMed  Google Scholar 

  37. Velie BD, Shrestha M, Franҫois L, Schurink A, Tesfayonas YG, Stinckens A, et al. Using an inbred horse breed in a high density genome-wide scan for genetic risk factors of insect bite hypersensitivity (IBH). PLoS ONE. 2016;11:e0152966.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Purcell S, Chang C. PLINK v1.90. 2021. https://www.cog-genomics.org/plink/1.9/ Accessed 11 February 2024.

  39. Purcell S. PLINK: a toolset for whole-genome association and population-based linkage analysis. Am J hum Genet. 2007;81:559–75.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Zheng X, Levine D, Shen J, Gogarten SM, Laurie C, Weir BS. A high-performance computing toolset for relatedness and principal component analysis of SNP data. Bioinformatics. 2012;28:3326–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. R Core team. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2019.

    Google Scholar 

  42. Wellman R, optiSel. Optimum Contribution Selection and Population Genetics. 2023. https://CRAN.R-project.org/package=optiSel. Accessed 11 February 2024.

  43. Barbato M, Orozco-terWengel P, Tapio M, Bruford MW. SNeP: a tool to estimate trends in recent effective population size trajectories using genome-wide SNP data. Front Genet. 2015;6:109.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Corbin LJ, Liu A, Bishop S, Woolliams J. Estimation of historical effective population size using linkage disequilibria with marker data. J Anim Breed Genet. 2012;129:257–70.

    Article  CAS  PubMed  Google Scholar 

  45. Sved J, Feldman M. Correlation and probability methods for one and two loci. Theor Popul Biol. 1973;4:129–32.

    Article  CAS  PubMed  Google Scholar 

  46. Biscarini F, Cozzi P, Gaspa G, Marras G, detectRUNS. Detect runs of homozygosity and runs of heterozygosity in diploid genomes. 2018. https://cran.r-project.org/package=detectRUNS. Accessed 11 February 2024.

  47. Grilz-Seger G, Mesarič M, Cotman M, Neuditschko M, Druml T, Brem G. Runs of homozygosity and population history of three horse breeds with small population size. J Equine Vet Sci. 2018;71:27–34.

    Article  Google Scholar 

  48. Nazari F, Seyedabadi H-R, Noshary A, Emamjomeh-Kashan N, Banabazi M-H. A genome-wide scan for signatures of selection in kurdish horse breed. J Equine Vet Sci. 2022;113:103916.

    Article  PubMed  Google Scholar 

  49. Amano T, Yokawa H, Masuda Y, Tozaki T, Kawai M, Shirai K. Genome-wide search reveals the uniqueness of DNA regions associated with coat color and innate immunity in Hokkaido native horse. Anim Sci J. 2023;94:e13884.

    Article  CAS  PubMed  Google Scholar 

  50. Cunningham F, Allen JE, Allen J, Alvarez-Jarreta J, Amode MR, Armean IM, et al. Ensembl 2022. Nucleic Acids Res. 2022;50:D988–95.

    Article  CAS  PubMed  Google Scholar 

  51. Thomas PD, Ebert D, Muruganujan A, Mushayahama T, Albou L-P, Mi H. PANTHER: making genome-scale phylogenetics accessible to all. Protein Sci. 2022;31:8–22.

    Article  CAS  PubMed  Google Scholar 

  52. Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, et al. The GeneCards suite: from gene data mining to disease genome sequence analyses. Curr Protoc Bioinform. 2016;54:1301–13033.

    Article  Google Scholar 

  53. Safran M, Rosen N, Twik M, BarShir R, Stein TI, Dahary D, et al. The GeneCards suite. In: Abugessaisa I, Kasukawa T, editors. Practical guide to Life Science Databases. Singapore: Springer Nature Singapore; 2021. pp. 27–56.

    Chapter  Google Scholar 

  54. Hu Z-L, Park CA, Reecy JM. Bringing the animal QTLdb and CorrDB into the future: meeting new challenges and providing updated services. Nucleic Acids Res. 2022;50:D956–61.

    Article  CAS  PubMed  Google Scholar 

  55. Schaefer RJ, Schubert M, Bailey E, Bannasch DL, Barrey E, Bar-Gal GK, et al. Developing a 670k genotyping array to tag ~ 2 M SNPs across 24 horse breeds. BMC Genomics. 2017;18:565.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Björnsson GB, Sveinsson HJ. Á spjöldum sögunnar [The history of the Icelandic horse]. In: Björnsson GB, Sveinsson HJ, editors. Íslenski hesturinn [The Icelandic horse]. 1st ed. Reykjavík: Mál og menning; 2006. pp. 76–101.

    Google Scholar 

  57. Björnsson GB, Sveinsson HJ. Á tímamótum [At a turning point in time]. In: Björnsson GB, Sveinsson HJ, editors. Íslenski hesturinn [The Icelandic horse]. 1st ed. Reykjavík: Mál og menning; 2006. pp. 102–13.

    Google Scholar 

  58. Árnason T, Klemetsdal G, Sigurssson Á. Nordiska hästraser - gamla genresurser ägnade for framtidens behov. In: International Symposium on Horse Breeding and Production in Cold Climatic Regions; 11–13 August; Hotel Saga. Reykjavík, Iceland; 1993.

  59. Kristjansson T. Erfðafjölbreytileiki íslenska hrossastofnsins og verndun hans. In: Dýrmundsson Ó, editor. Fjölrit LbhÍ Nr. 14. Íslensk búfjárrækt: Málstofa til heiðurs Hjalta Gestssyni níræðum. 2006 November 17; Reykjavik, Iceland. Hvanneyri, Iceland: The Agricultural University of Iceland; 2007. pp. 95–102.

    Google Scholar 

  60. Animal Importation. Act 1990 (IS) s 2.

  61. Sumreddee P, Hay EH, Toghiani S, Roberts A, Aggrey SE, Rekaya R. Grid search approach to discriminate between old and recent inbreeding using phenotypic, pedigree and genomic information. BMC Genomics. 2021;22:538.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Stoffel MA, Johnston SE, Pilkington JG, Pemberton JM. Mutation load decreases with haplotype age in wild Soay sheep. Evol Lett. 2021;5:187–95.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Ansari HA, Hediger R, Fries R, Stranzinger G. Chromosomal localization of the major histocompatibility complex of the horse (ELA) by in situ hybridization. Immunogenetics. 1988;28:362–4.

    Article  CAS  PubMed  Google Scholar 

  64. Gustafson A, Tallmadge RL, Ramlachan N, Miller D, Bird H, Antczak DF, et al. An ordered BAC contig map of the equine major histocompatibility complex. Cytogenet Genome Res. 2003;102:189–95.

    Article  CAS  PubMed  Google Scholar 

  65. Kelley J, Walter L, Trowsdale J. Comparative genomics of major histocompatibility complexes. Immunogenetics. 2005;56:683–95.

    Article  CAS  PubMed  Google Scholar 

  66. Holmes CM, Violette N, Miller D, Wagner B, Svansson V, Antczak DF. MHC haplotype diversity in Icelandic horses determined by polymorphic microsatellites. Genes Immun. 2019;20:660–70.

    Article  PubMed  Google Scholar 

  67. Solé M, Ablondi M, Binzer-Panchal A, Velie BD, Hollfelder N, Buys N, et al. Inter- and intra-breed genome-wide copy number diversity in a large cohort of European equine breeds. BMC Genomics. 2019;20:759.

    Article  PubMed  PubMed Central  Google Scholar 

  68. Laseca N, Molina A, Valera M, Antonini A, Demyda-Peyrás S. Copy number variation (CNV): a new genomic insight in horses. Animals. 2022;12:1435.

    Article  PubMed  PubMed Central  Google Scholar 

  69. Wang W, Wang S, Hou C, Xing Y, Cao J, Wu K, et al. Genome-wide detection of copy number variations among diverse horse breeds by array CGH. PLoS ONE. 2014;9:e86860.

    Article  PubMed  PubMed Central  Google Scholar 

  70. Redon R, Ishikawa S, Fitch KR, Feuk L, Perry GH, Andrews TD, et al. Global variation in copy number in the human genome. Nature. 2006;444:444–54.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. Andersson LS, Larhammar M, Memic F, Wootz H, Schwochow D, Rubin C-J, et al. Mutations in DMRT3 affect locomotion in horses and spinal circuit function in mice. Nature. 2012;488:642–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  72. Kristjansson T, Bjornsdottir S, Sigurdsson A, Andersson L, Lindgren G, Helyar S, et al. The effect of the ‘Gait keeper’ mutation in the DMRT3 gene on gaiting ability in Icelandic horses. J Anim Breed Genet. 2014;131:415–25.

    Article  CAS  PubMed  Google Scholar 

  73. Velie BD, Fegraeus KJ, Solé M, Rosengren MK, Røed KH, Ihler C-F, et al. A genome-wide association study for harness racing success in the norwegian-swedish coldblooded trotter reveals genes for learning and energy metabolism. BMC Genet. 2018;19:80.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  74. Jäderkvist K, Andersson LS, Johansson AM, Árnason T, Mikko S, Eriksson S, et al. The DMRT3 ‘gait keeper’ mutation affects performance of nordic and standardbred trotters. J Anim Sci. 2014;92:4279–86.

    Article  PubMed  Google Scholar 

  75. Promerová M, Andersson L, Juras R, Penedo M, Reissmann M, Tozaki T, et al. Worldwide frequency distribution of the ‘Gait keeper’ mutation in the DMRT3 gene. Anim Genet. 2014;45:274–82.

    Article  PubMed  Google Scholar 

  76. Patterson L, Staiger E, Brooks S. DMRT3 is associated with gait type in Mangalarga Marchador horses, but does not control gait ability. Anim Genet. 2015;46:213–5.

    Article  CAS  PubMed  Google Scholar 

  77. Novoa-Bravo M, Jäderkvist Fegraeus K, Rhodin M, Strand E, García LF, Lindgren G. Selection on the Colombian paso horse’s gaits has produced kinematic differences partly explained by the DMRT3 gene. PLoS ONE. 2018;13:e0202584.

    Article  PubMed  PubMed Central  Google Scholar 

  78. Chandra Paul R, Ba Nguyen T, Okuda Y, Nu Anh Le T, Mosese Dau Tabuyaqona J, Konishi Y, et al. Distribution of the mutant allele of the DMRT3 gene associated with ambling gaits in Japanese native horse populations. Anim Sci J. 2020;91:e13431.

    Article  CAS  PubMed  Google Scholar 

  79. Staiger EA, Almén MS, Promerová M, Brooks S, Cothran EG, Imsland F, et al. The evolutionary history of the DMRT3 ‘Gait keeper’ haplotype. Anim Genet. 2017;48:551–9.

    Article  CAS  PubMed  Google Scholar 

  80. Bas Conn L. The role of polymorphisms of the MSTN, GRIN2B and DOCK8 genes in the performance of pace-racing Icelandic horses. MSc thesis, Swedish University of Agricultural Sciences. 2018.

  81. Kang JU, Koo SH, Kwon KC, Park JW. Frequent silence of chromosome 9p, homozygous DOCK8, DMRT1 and DMRT3 deletion at 9p24. 3 in squamous cell carcinoma of the lung. Int J Oncol. 2010;37:327–35.

    CAS  PubMed  Google Scholar 

  82. Glessner JT, Li J, Wang D, March M, Lima L, Desai A, et al. Copy number variation meta-analysis reveals a novel duplication at 9p24 associated with multiple neurodevelopmental disorders. Genome Med. 2017;9:106.

    Article  PubMed  PubMed Central  Google Scholar 

  83. Jäderkvist K, Holm N, Imsland F, Árnason T, Andersson L, Andersson LS, et al. The importance of the DMRT3 ‘Gait keeper’ mutation on riding traits and gaits in Standardbred and Icelandic horses. Livest Sci. 2015;176:33–9.

    Article  Google Scholar 

  84. Molt S, Bührdel JB, Yakovlev S, Schein P, Orfanos Z, Kirfel G, et al. Aciculin interacts with filamin C and Xin and is essential for myofibril assembly, remodeling and maintenance. J Cell Sci. 2014;127:3578–92.

    CAS  PubMed  Google Scholar 

  85. Metallinos DL, Bowling AT, Rine J. A missense mutation in the endothelin-B receptor gene is associated with lethal white foal syndrome: an equine version of Hirschsprung disease. Mamm Genome. 1998;9:426–31.

    Article  CAS  PubMed  Google Scholar 

  86. Santschi EM, Purdy AK, Valberg SJ, Vrotsos PD, Kaese H, Mickelson JR. Endothelin receptor B polymorphism associated with lethal white foal syndrome in horses. Mamm Genome. 1998;9:306–9.

    Article  CAS  PubMed  Google Scholar 

  87. Yan GC, Croaker D, Zhang AL, Manglick P, Cartmill T, Cass D. A dinucleotide mutation in the endothelin-B receptor gene is associated with lethal white foal syndrome (LWFS); a horse variant of Hirschsprung disease (HSCR). Hum Mol Genet. 1998;7:1047–52.

    Article  Google Scholar 

  88. Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, et al. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature. 1988;332:411–5.

    Article  CAS  PubMed  Google Scholar 

  89. Inoue A, Yanagisawa M, Kimura S, Kasuya Y, Miyauchi T, Goto K, et al. The human endothelin family: three structurally and pharmacologically distinct isopeptides predicted by three separate genes. P Natl Acad Sci USA. 1989;86:2863–7.

    Article  CAS  Google Scholar 

  90. Baynash AG, Hosoda K, Giaid A, Richardson JA, Emoto N, Hammer RE, et al. Interaction of endothelin-3 with endothelin-B receptor is essential for development of epidermal melanocytes and enteric neurons. Cell. 1994;79:1277–85.

    Article  CAS  PubMed  Google Scholar 

  91. Hosoda K, Hammer RE, Richardson JA, Baynash AG, Cheung JC, Giaid A, et al. Targeted and natural (piebald-lethal) mutations of endothelin-B receptor gene produce megacolon associated with spotted coat color in mice. Cell. 1994;79:1267–76.

    Article  CAS  PubMed  Google Scholar 

  92. Stanchina L, Baral V, Robert F, Pingault V, Lemort N, Pachnis V, et al. Interactions between Sox10, Edn3 and Ednrb during enteric nervous system and melanocyte development. Dev Biol. 2006;295:232–49.

    Article  CAS  PubMed  Google Scholar 

  93. Jäderkvist Fegraeus K, Velie BD, Axelsson J, Ang R, Hamilton NA, Andersson L, et al. A potential regulatory region near the EDN3 gene may control both harness racing performance and coat color variation in horses. Physiol Rep. 2018;6:e13700.

    Article  PubMed  PubMed Central  Google Scholar 

  94. Fegraeus K, Rosengren MK, Naboulsi R, Orlando L, Åbrink M, Jouni A, et al. An endothelial regulatory module links blood pressure regulation with elite athletic performance. PLOS Genet. 2024;20:e1011285.

    Article  PubMed  PubMed Central  Google Scholar 

  95. Stefánsdóttir G, Ragnarsson S, Gunnarsson V, Jansson A. Physiological response to a breed evaluation field test in Icelandic horses. Animal. 2014;8:431–9.

    Article  PubMed  PubMed Central  Google Scholar 

  96. Stefánsdóttir G, Ragnarsson S, Gunnarsson V, Roepstorff L, Jansson A. A comparison of the physiological response to tölt and trot in the Icelandic horse. J Anim Sci. 2015;93:3862–70.

    Article  PubMed  Google Scholar 

  97. Hirst CE, Lim S-M, Pereira LA, Mayberry RA, Stanley EG, Elefanty AG. Expression from a betageo gene trap in the Slain1 gene locus is predominantly associated with the developing nervous system. Int J Dev Biol. 2010;54:1383–8.

    Article  CAS  PubMed  Google Scholar 

  98. Pavlopoulos E, Trifilieff P, Chevaleyre V, Fioriti L, Zairis S, Pagano A, et al. Neuralized1 activates CPEB3: a function for nonproteolytic ubiquitin in synaptic plasticity and memory storage. Cell. 2011;147:1369–83.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  99. Taal K, Tuvikene J, Rullinkov G, Piirsoo M, Sepp M, Neuman T, et al. Neuralized family member NEURL1 is a ubiquitin ligase for the cGMP-specific phosphodiesterase 9A. Sci Rep. 2019;9:7104.

    Article  PubMed  PubMed Central  Google Scholar 

  100. Rieder S, Taourit S, Mariat D, Langlois B, Guérin G. Mutations in the agouti (ASIP), the extension (MC1R), and the brown (TYRP1) loci and their association to coat color phenotypes in horses (Equus caballus). Mamm Genome. 2001;12:450–5.

    Article  CAS  PubMed  Google Scholar 

  101. Imsland F, McGowan K, Rubin C-J, Henegar C, Sundström E, Berglund J, et al. Regulatory mutations in TBX3 disrupt asymmetric hair pigmentation that underlies Dun camouflage color in horses. Nat Genet. 2016;48:152–8.

    Article  CAS  PubMed  Google Scholar 

  102. Lee S-T, Nicholls RD, Jong MT, Fukai K, Spritz RA. Organization and sequence of the human P gene and identification of a new family of transport proteins. Genomics. 1995;26:354–63.

    Article  CAS  PubMed  Google Scholar 

  103. Sturm RA, Frudakis TN. Eye colour: portals into pigmentation genes and ancestry. Trends Genet. 2004;20:327–32.

    Article  CAS  PubMed  Google Scholar 

  104. Sturm RA, Duffy DL, Zhao ZZ, Leite FP, Stark MS, Hayward NK, et al. A single SNP in an evolutionary conserved region within intron 86 of the HERC2 gene determines human blue-brown eye color. Am J Hum Genet. 2008;82:424–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  105. Duffy DL, Montgomery GW, Chen W, Zhao ZZ, Le L, James MR, et al. A three-single-nucleotide polymorphism haplotype in intron 1 of OCA2 explains most human eye-color variation. Am J Hum Genet. 2007;80:241–52.

    Article  CAS  PubMed  Google Scholar 

  106. Sturm RA. Molecular genetics of human pigmentation diversity. Hum Mol Genet. 2009;18:R9–17.

    Article  CAS  PubMed  Google Scholar 

  107. Kowalski EJA, Bellone RR. Investigation of HERC2 and OCA2 SNP for iris color variation in Puerto Rican Paso Fino horses. J Equine Vet Sci. 2011;31:319.

    Article  Google Scholar 

  108. Bellone R, Lawson S, Hunter N, Archer S, Bailey E. Analysis of a SNP in exon 7 of equine OCA2 and its exclusion as a cause for Appaloosa spotting. Anim Genet. 2006;37:525.

    Article  CAS  PubMed  Google Scholar 

  109. General rules and regulations: Breeding rules and regulations. 2023. https://www.feiffengur.com/documents/FEIF Breeding_2023.pdf. Accessed 5 January 2024.

  110. Zhang T, Kim DH, Xiao X, Lee S, Gong Z, Muzumdar R, et al. FoxO1 plays an important role in regulating β-cell compensation for insulin resistance in male mice. Endocrinology. 2016;157:1055–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  111. Puig O, Tjian R. Transcriptional feedback control of insulin receptor by dFOXO/FOXO1. Gene Dev. 2005;19:2435–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  112. Matsumoto M, Han S, Kitamura T, Accili D. Dual role of transcription factor FoxO1 in controlling hepatic insulin sensitivity and lipid metabolism. J Clin Invest. 2006;116:2464–72.

    CAS  PubMed  PubMed Central  Google Scholar 

  113. Frank N, Geor RJ, Bailey SR, Durham AE, Johnson PJ. Equine metabolic syndrome. J Vet Intern Med. 2010;24:467–75.

    Article  CAS  PubMed  Google Scholar 

  114. Bröjer J, Lindåse S, Hedenskog J, Alvarsson K, Nostell K. Repeatability of the combined glucose-insulin tolerance test and the effect of a stressor before testing in horses of 2 breeds. J Vet Intern Med. 2013;27:1543–50.

    Article  PubMed  Google Scholar 

  115. Bailey SR, Habershon-Butcher JL, Ransom KJ, Elliott J, Menzies-Gow NJ. Hypertension and insulin resistance in a mixed-breed population of ponies predisposed to laminitis. Am J Vet Res. 2008;69:122–9.

    Article  CAS  PubMed  Google Scholar 

  116. Jansson P-A. Endothelial dysfunction in insulin resistance and type 2 diabetes. J Vet Intern Med. 2007;262:173–83.

    CAS  Google Scholar 

  117. Johnson P. The equine metabolic syndrome peripheral Cushing’s syndrome. Vet Clin N Am-Equine. 2002;18:271–93.

    Article  Google Scholar 

  118. Gieger C, Radhakrishnan A, Cvejic A, Tang W, Porcu E, Pistis G, et al. New gene functions in megakaryopoiesis and platelet formation. Nature. 2011;480:201–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  119. Qayyum R, Snively BM, Ziv E, Nalls MA, Liu Y, Tang W, et al. A meta-analysis and genome-wide association study of platelet count and mean platelet volume in African americans. PLOS Genet. 2012;8:e1002491.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  120. Freedman BI, Bowden DW, Ziegler JT, Langefeld CD, Lehtinen AB, Rudock ME, et al. Bone morphogenetic protein 7 (BMP7) gene polymorphisms are associated with inverse relationships between vascular calcification and BMD: the diabetes heart study. J Bone Min Res. 2009;24:1719–27.

    Article  CAS  Google Scholar 

  121. Lumbroso S, Paris Fo, Sultan C. Activating Gsα mutations: analysis of 113 patients with signs of McCune-Albright syndrome - A European collaborative study. J Clin Endocr Metab. 2004;89:2107–13.

    Article  CAS  PubMed  Google Scholar 

  122. Szmatoła T, Gurgul A, Jasielczuk I, Oclon E, Ropka-Molik K, Stefaniuk-Szmukier M, et al. Assessment and distribution of runs of homozygosity in horse breeds representing different utility types. Animals. 2022;12:3293.

    Article  PubMed  PubMed Central  Google Scholar 

  123. Makvandi-Nejad S, Hoffman GE, Allen JJ, Chu E, Gu E, Chandler AM, et al. Four loci explain 83% of size variation in the horse. PLoS ONE. 2012;7:e39929.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  124. Bartholazzi Junior A, Quirino CR, Vega WHO, Rua MAS, David CMG, Jardim JG. Polymorphisms in the LASP1 gene allow selection for smaller stature in ponies. Livest Sci. 2018;216:160–4.

    Article  Google Scholar 

  125. Skujina I, Winton CL, Hegarty MJ, McMahon R, Nash DM, Morel MCGD, et al. Detecting genetic regions associated with height in the native ponies of the British Isles by using high density SNP genotyping. Genome. 2018;61:767–70.

    Article  CAS  PubMed  Google Scholar 

  126. Thomer A, Gottschalk M, Christmann A, Naccache F, Jung K, Hewicker-Trautwein M, et al. An epistatic effect of KRT25 on SP6 is involved in curly coat in horses. Sci Rep. 2018;8:6374.

    Article  PubMed  PubMed Central  Google Scholar 

  127. Yuki KE, Marei H, Fiskin E, Eva MM, Gopal AA, Schwartzentruber JA, et al. CYRI/FAM49B negatively regulates RAC1-driven cytoskeletal remodelling and protects against bacterial infection. Nat Microbiol. 2019;4:1516–31.

    Article  CAS  PubMed  Google Scholar 

  128. Van den Eede A, Martens A, Lipinska U, Struelens M, Deplano A, Denis O, et al. High occurrence of methicillin-resistant Staphylococcus aureus ST398 in equine nasal samples. Vet Microbiol. 2009;133:138–44.

    Article  PubMed  Google Scholar 

  129. Sweeney CR, Timoney JF, Newton JR, Hines MT. Streptococcus equi infections in horses: guidelines for treatment, control, and prevention of strangles. J Vet Intern Med. 2005;19:123–34.

    Article  PubMed  Google Scholar 

  130. Uzal FA, Navarro MA, Asin J, Henderson EE. Clostridial diseases of horses: a review. Vaccines. 2022;10:318.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  131. Warner SL, Boggs J, Lee JK, Reddy S, Banes M, Cooley J. Clinical, pathological, and genetic characterization of Listeria monocytogenes causing sepsis and necrotizing typhlocolitis and hepatitis in a foal. J Vet Diagn Invest. 2012;24:581–6.

    Article  PubMed  Google Scholar 

  132. Irwin DM, Biegel JM, Stewart C-B. Evolution of the mammalian lysozyme gene family. BMC Evol Biol. 2011;11:166.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  133. Mastrangelo S, Tolone M, Sardina MT, Sottile G, Sutera AM, Di Gerlando R, et al. Genome-wide scan for runs of homozygosity identifies potential candidate genes associated with local adaptation in Valle Del Belice sheep. Genet Sel Evol. 2017;49:84.

    Article  PubMed  PubMed Central  Google Scholar 

  134. Greenbaum MP, Yan W, Wu M-H, Lin Y-N, Agno JE, Sharma M, et al. TEX14 is essential for intercellular bridges and fertility in male mice. P Natl Acad Sci-Biol. 2006;103:4982–7.

    Article  CAS  Google Scholar 

  135. Greenbaum MP, Iwamori N, Agno JE, Matzuk MM. Mouse TEX14 is required for embryonic germ cell intercellular bridges but not female fertility. Biol Reprod. 2009;80:449–57.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  136. Baudat F, Manova K, Yuen JP, Jasin M, Keeney S. Chromosome synapsis defects and sexually dimorphic meiotic progression in mice lacking Spo11. Mol Cell. 2000;6:989–98.

    Article  CAS  PubMed  Google Scholar 

  137. Carrell D, De Jonge C, Lamb D. The genetics of male infertility: a field of study whose time is now. Arch Androl. 2006;52:269–74.

    Article  CAS  PubMed  Google Scholar 

  138. Zhang J, Zhou D-x, Wang H-x, Tian Z. An association study of SPO11 gene single nucleotide polymorphisms with idiopathic male infertility in Chinese Han population. J Assist Reprod Gen. 2011;28:731–6.

    Article  Google Scholar 

  139. Ghalkhani E, Sheidai M, Gourabi H, Noormohammadi Z, Bakhtari N, Malekasgar AM. Study of single nucleotide polymorphism (rs28368082) in SPO11 gene and its association with male infertility. J Assist Reprod Gen. 2014;31:1205–10.

    Article  Google Scholar 

  140. Nicolini P, Amorín R, Han Y, Peñagaricano F. Whole-genome scan reveals significant non-additive effects for sire conception rate in Holstein cattle. BMC Genet. 2018;19:14.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  141. Sha Y-W, Xu X, Ji Z-Y, Mei L-B, Qiu P-P, Ji H, et al. Sperm-egg fusion disorder in a Chinese male patient was associated with a rare ADAM20 variant. Oncotarget. 2017;9:2086–91.

    Article  PubMed  PubMed Central  Google Scholar 

  142. Zhu G-Z, Lin Y, Myles DG, Primakoff P. Identification of four novel ADAMs with potential roles in spermatogenesis and fertilization. Gene. 1999;234:227–37.

    Article  CAS  PubMed  Google Scholar 

  143. vanH RH. ADAM 20 and 21; two novel human testis-specific membrane metalloproteases with similarity to fertilin-α. Gene. 1998;206:273–82.

  144. Li Y, Chen Y, Wu W, Li N, Hua J. MMPs, ADAMs and ADAMTSs are associated with mammalian sperm fate. Theriogenology. 2023;200:147–54.

    Article  CAS  PubMed  Google Scholar 

  145. Cezard T, Cunningham F, Hunt SE, Koylass B, Kumar N, Saunders G, Shen A, Silva AF, Tsukanov K, Venkataraman S, Flicek P, Parkinson H, Keane TM. The European variation archive: a FAIR resource of genomic variation for all species. Nucleic Acids Res. 2021;50:D1216–20.

    Article  PubMed Central  Google Scholar 

  146. Percie du Sert N, Hurst V, Ahluwalia A, Alam S, Avey MT, Baker M, et al. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLOS Biol. 2020;18:e3000410.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors extend their gratitude to all horse owners for their cooperation and contributions. Special thanks are due to Elsa Albertsdóttir for providing valuable insights and pedigree data for the Icelandic horse, to Marie Rhodin for engaging discussions, to Tytti Vanhala at the Swedish University of Agricultural Sciences for her work in DNA extraction and genotyping, and to Diana Ekman of the Swedish Bioinformatics Advisory Program at SciLifeLab for her expertise and guidance in bioinformatics. The publication was made with the contribution of the researcher Michela Ablondi which is employed with a research contract co-financed by the European Union – PON Research and Innovation 2014-2020 pursuant to art. 24, paragraph 3, lett. a) of Law 30 December 2010, n. 240 and subsequent amendments and of the D.M. 10 August 2021 no. 1062.

Funding

Open access funding provided by Swedish University of Agricultural Sciences. The Agricultural University of Iceland’s Doctoral Fund, The Stock Protection Fund for the Icelandic Horse Breed, The Swedish Icelandic Horse Association (SIF), Blikastaðir Fund, The Icelandic Genetic Resource Council in Agriculture (AGROGEN) and The Swedish Research Council (VR).

Open access funding provided by Swedish University of Agricultural Sciences.

Author information

Authors and Affiliations

  1. Department of Animal Biosciences, Swedish University of Agricultural Sciences, P.O. Box 7023, Uppsala, 75007, Sweden

    Heiðrún Sigurðardóttir, Gabriella Lindgren & Susanne Eriksson

  2. Faculty of Agricultural Sciences, Agricultural University of Iceland, Hvanneyri, Borgarbyggð, 311, Iceland

    Heiðrún Sigurðardóttir & Thorvaldur Kristjansson

  3. Department of Veterinary Science, University of Parma, Parma, 43126, Italy

    Michela Ablondi

  4. Center for Animal Breeding and Genetics, Department of Biosystems, KU Leuven, Leuven, 3001, Belgium

    Gabriella Lindgren

Authors
  1. Heiðrún Sigurðardóttir
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Michela Ablondi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Thorvaldur Kristjansson
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Gabriella Lindgren
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Susanne Eriksson
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

HS, TK, GL, and SE initiated and designed the study. HS collected half of the samples. MA provided scripts for the data analysis. HS performed the experiments and data analysis and drafted the manuscript. SE and MA advised with the statistical analysis. HS, MA, TK, GL, and SE contributed to the interpretation of the results. All authors read, suggested modifications, and approved the final manuscript.

Corresponding author

Correspondence to Heiðrún Sigurðardóttir.

Ethics declarations

Ethics approval and consent to participate

This study is reported in accordance with ARRIVE guidelines [146]. Hair samples were collected according to ethical approval by the Ethics Committee for Animal Experiments in Uppsala, Sweden (number: 5.8.18–15453/2017) and an animal experiment license by the Icelandic Food and Veterinary Authority in Iceland (number: 2020-04-02/2003120). The study involved only privately owned horses, and informed consent was obtained from the horse owners for their participation in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare competing interests concerning the commercial applications of the current study. GL is a co-inventor of a patent application concerning commercial testing of the DMRT3 mutation. The stated patent does not restrict research applications of the method. None of the other authors have any competing interests.

Authors’ information

Not applicable.

Footnotes

Not applicable.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sigurðardóttir, H., Ablondi, M., Kristjansson, T. et al. Genetic diversity and signatures of selection in Icelandic horses and Exmoor ponies. BMC Genomics 25, 772 (2024). https://doi.org/10.1186/s12864-024-10682-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12864-024-10682-8

Keywords

BMC Genomics

ISSN: 1471-2164

Contact us