Journal of Equine Veterinary Science 33 (2013) 244-249 Journal of Equine Veterinary Science journal homepage: www.j -evs.com Original Research Morphological and Genomic Differences Between Cutting and Racing Lines of Quarter Horses Camila Tangari Meira MD a, Rogério Abdallah Curi PhD b, Josineudson Augusto II Vasconcelos Silva PhD b, Marcio José Monteiro Corrêa b, Henrique Nunes de Oliveira a, Marcilio Dias Silveira da Mota PhD b a Faculdade de Ciências Agrárias e Veterinárias, Universidade Estadual Paulista, Jaboticabal, Sao Paulo, Brazil b Faculdade de Medicina Veterinária e Zootecnia, Universidade Estadual Paulista, Botucatu, Sao Paulo, Brazil a r t i c l e i n f o Article history: Received 2 April 2012 Received in revised form 11 June 2012 Accepted 11 July 2012 Available online 15 September 2012 Keywords: Physical traits Horse Genome-wide association study SNPs Corresponding author at: Camila Tangari Mei de Ciências Agrárias e Veterinárias, Universidade Jaboticabal, Sao Paulo 14884-900, Brazil. E-mail address: camilatangari@zootecnista.com.br 0737-0806 � 2013 Elsevier Inc. http://dx.doi.org/10.1016/j.jevs.2012.07.001 Open access under the E a b s t r a c t To investigate morphological and genomic differences between cutting and racing lines of Quarter Horses, 120 racing and 68 cutting animals of both sexes, registered at the Brazilian Association of Quarter Horse Breeders, were used. Blood samples were collected, and the following physical traits were measured: weight; height at withers; body length; length of the shank, pastern, rump, head, and neck; and chest, shank, and hoof circumference. For analysis of genomic differences, 54,602 single-nucleotide poly- morphisms (SNPs) were genotyped using the Equine SNP50 BeadChip, and the quality of individual and SNP genotype data were evaluated. The fixation index, FST, was used to identify genome regions that were altered in the lines by selection. The results showed significant differences between the lines in all physical traits. Quality control led to the exclusion of four cutting animals with a call rate of <0.95. After filtering, 12,544, 13,815, and 13,370 SNPs were excluded for the whole population (n ¼ 184), the 120 racing animals, and the 64 cutting animals, respectively. The number of informative poly- morphisms detected in each line and in the whole population indicated that the Equine SNP50 BeadChip can be used in genetic studies of Quarter Horses. The fixation index, FST, identified 2,558 genome regions that may have been modified by divergent selection. � 2013 Elsevier Inc. Open access under the Elsevier OA license. 1. Introduction As a breed of global importance, corresponding to 52% of all horses, Quarter Horses are important because of their great versatility in different equestrian events [1]. Quarter Horses were developed in North America in the 17th century from Arabian and Turkish horses brought by European settlers. The major development of this breed occurred during westward expansion when pioneers needed robust and versatile horses, fit for the saddle and ra, MD, Faculdade Estadual Paulista, (C.T. Meira). lsevier OA license. for traction, in viewof the difficulty to keep a varied stock of animals to satisfy diverse necessities [1]. The Quarter Horse breed is subdivided into different lines according to skills resulting from distinct selection objectives, including cutting and racing horses. The cutting line is destined for functional tests, exploring skills such as agility and obedience, which are important for cattle management in the field. The racing line is characterized by great sprinting speed over short distances on straight tracks. The cutting type is shorter and more compact and hasmuscular hindquarters, whereas the racing type is taller and has longer legs and a less prominent musculature. The simultaneous study of thousands of DNA poly- morphisms spread across the genome, known as genome- wide association analysis, has permitted the study of Delta:1_given name Delta:1_given name Delta:1_surname Delta:1_given name Delta:1_given name Delta:1_surname Delta:1_given name Delta:1_given name Delta:1_surname Delta:1_given name Delta:1_given name Delta:1_surname mailto:camilatangari@zootecnista.com.br http://www.j-evs.com http://dx.doi.org/10.1016/j.jevs.2012.07.001 http://dx.doi.org/10.1016/j.jevs.2012.07.001 http://www.elsevier.com/open-access/userlicense/1.0/ http://www.elsevier.com/open-access/userlicense/1.0/ C.T. Meira et al. / Journal of Equine Veterinary Science 33 (2013) 244-249 245 different populations of various domestic animal species [2-4], as well as the estimation of genetic divergencewithin and between populations [5]. During the process of domestication and breed formation, domestic animals were subjected to natural and artificial selection. These selection pressures led to an increase in the frequency of some mutations in specific regions of the genome, which generated more adapted animals or provided them with favorable characteristics tomeet human needs. At the same time, the frequency of other polymorphisms decreased or they were completely eliminated [6]. In this respect, the comparison of allele frequencies between selected and unselected populations or between populations selected for different objectives provides insights into the regions of the genome that have been modified by selection. According to Chowdhary and Raudsepp [7], one of the highlights from the analysis of the horse genome is its complete sequencing from a Thoroughbred animal (Equ- Cab2.0) and, from this, the identification of 1,162,753 single-nucleotide polymorphisms (SNPs) in different breeds [8]. Designed to identify SNPs and genes that contribute to traits of interest in the major horse breeds raised today in the world, the Equine SNP50 BeadChip developed by Illumina, Inc., (San Diego, California, USA) represents a powerful platform for genetic studies of this species, permitting researchers to perform a variety of experiments that require the genotyping of DNA polymorphisms. In view of these considerations, the aims of the present study were to investigate the genomic differences, using the Equine SNP50 BeadChip, andmorphological differences between cutting and racing lines of Quarter Horses as a result of selection for different objectives. 2. Materials and Methods 2.1. Animals and Phenotypic Data One hundred eighty-eight Quarter Horses of both sexes born between 1985 and 2009, including 120 racing horses and 68 cutting horses, registered at the Brazilian Associa- tion of Quarter Horse Breeders, were studied. All experi- mental procedures were conducted in accordance with the Brazilian legislation on animal welfare. The following physical traits were measured according to Torres and Jardim [9]: weight; height at withers; body length; length of the shank, pastern, rump, head, and neck; and chest, shank, and hoof circumference. The measure- ments were performed by the same person using a tape measure andmeasuring stick, always on the right side of the animal, with the horse standing with front and rear legs perpendicular to the ground. For genotyping, a 5-mL sample of whole bloodwas collected from each animal by puncture of the left jugular vein in the neck region into vacuum tubes containing 7.5 mg ethylenediaminetetraacetic acid. The animals of the racing line (18 male and 102 female horses), born to 48 stallions and 107 mares, belonged to five farms in the countryside of the State of São Paulo, Brazil. The animals of the cutting line (26 male and 42 female horses), born to 44 stallions and 64 mares, belonged to three other farms in the countryside of the State of São Paulo. In both lines, full sibs were avoided. 2.2. Genotyping of SNPs Genomic DNA was extracted from the blood samples of Quarter Horses using the Illustra Blood Genomicprep Mini Spin kit (GE Healthcare, Little Chalfont, Buckinghamshire, UK) according to manufacturer instructions. DNA integrity was analyzed using 0.8% agarose gel electrophoresis, and DNA was quantified using a NanoDrop ND-2000 spectro- photometer (Thermo Fisher Scientific, Waltham, Massa- chusetts, USA). The DNA concentration in the samples was adjusted to 40-60 ng/mL. SNPs were genotyped on the HiScan system (Illumina Inc.) using the Illumina Equine SNP50 BeadChip at the Faculty of Agricultural and Veterinary Sciences, UNESP, Jaboticabal, São Paulo, Brazil. The chip contains 54,602 SNPs uniformly distributed across the entire genome of 15 horse breeds. The SNPs are distributed across the 31 auto- somes and X chromosome. The mean interval between SNPs is 43,200 bp. This content is derived from the Equ- Cab2.0 SNP Collection compiled by the Broad Institute’s Equine Genome Sequencing Project, which identified >940,000 SNPs in Arabian, Andaluz, Akhal-Teke, Islandesa, Standardbred, Thoroughbred, and Quarter Horses. 2.3. Analysis of Morphological Differences and Differences in Inbreeding between Cutting and Racing Lines Morphological differences between the two Quarter Horse lines (cutting and racing) were evaluated using a model that included the effects of sex and line, and age at recording as covariate. The general linear model (GLM) procedure of the Statistical Analysis System v.9.1 program (SAS Institute Inc., Cary, North Carolina, USA) [10] was used for statistical analysis, andmeans were compared using the Tukey test at a level of significance of 5%. The inbreeding coefficient was calculated for each animal of the two lines based on pedigree records using the Relax2 program (MTT Agrifood Research Finland, Biomet- rical Genetics, Jokioinen, Finland) [11]. The relationship matrix contained 762 animals, with a depth of ancestry of four generations. The average inbreeding of consanguin- eous animals and of all animals of each line was estimated using the coefficients of inbreeding. 2.4. Quality of Genotype Data The quality of individual and SNP genotype data was investigated using the Genome Studio program, version 2011.1 (Illumina Inc.). For individuals, call rate, heterozy- gosity, and gender estimation were determined. Animals with a call rate <0.95, heterozygosity of �3 standard deviations from the mean, and errors in gender estimation were excluded from the sample. In addition, agreement between four replicates and parentage concordance (allele sharing) between four stallion/progeny and three stallion/ mare/progeny pairs were evaluated. With respect to the quality of SNP genotypes in the whole population and in each line, SNPs located on the X chromosome were excluded (filtered). SNPs with low gen- otyping quality (cluster separation <0.3), a call frequency <0.9, aminor allele frequency (MAF)<0.05 (including fixed alleles), and aHardyeWeinberg P< .001were also excluded. C.T. Meira et al. / Journal of Equine Veterinary Science 33 (2013) 244-249246 2.5. Analysis of Genomic Differences between Cutting and Racing Lines The fixation index, FST, was used to identify genome regions in the cutting and racing lines that have been modified by selection [12-14]. The w ¼ FST parameter was estimated for all SNPs that passed quality control (consid- ering all individuals) using the Bayesian method proposed by Gianola et al. [15]. This method is a two-step procedure that eventually leads to clusters of w values. The first step uses a simple Bayesian structure for removing samples from the posterior distribution of w parameters without constructing Markov chains. This step assigns a weakly informative prior of allele frequencies and does not make any assumptions about evolutionary models. The second step considers samples from this posterior distribution as “data” and fits a sequence of finite mixture models to identify clusters of w-statistics. 3. Results and Discussion Selecting Quarter Horses for different objectives (racing or cutting) promoted significant changes in the physical traits of the animals. Table 1 compares the fitted means of the characteristics analyzed between the two lines. A significant effect of line (P < .0001) was observed for all traits. As expected, racing animals presented higher weight, height, body length, and body circumference than cutting animals. Thus, the phenotypic differences observed between the two lines have a possible genetic background. However, it is necessary to consider that phenotypes are also influenced by several environmental factors, such as feeding, training, and others. Although there are no other morphological studies characterizing Quarter Horse lines, the present investigation is important because it demon- strated a close relationship between physical characteris- tics and the function for which the animal is used. The different abilities of the breed, such as running, jumping, reining, and barrel racing, have resulted in an appropriate biotype for each modality, and differences in performance exist owing to morphological adequacy or inadequacy. Among the 188 animals studied, 12 racing animals and one cutting animal were consanguineous. The mean coef- ficient of inbreeding (F) was 0.0028 or 0.28% for the whole population. In contrast, the mean coefficient of inbreeding Table 1 Least-square means and standard errors of physical traits in racing and cutting Quarter Horses Traits Racing Line Cutting Line Body length (m) 1.8081 � 0.0057a 1.6343 � 0.0087b Shank length (cm) 27.065 � 0.0014a 24.478 � 0.0014b Pastern length (cm) 13.382 � 0.0009a 12.231 � 0.0009b Rump length (cm) 62.032 � 0.0032a 55.173 � 0.0028b Head length (cm) 62.170 � 0.0020a 56.057 � 0.0021b Neck length (cm) 79.772 � 0.0034a 73.333 � 0.0049b Weight (kg) 538.97 � 3.7671a 450.69 � 5.6902b Height at withers (m) 1.5592 � 0.0040a 1.4660 � 0.0045b Chest circumference (m) 1.9669 � 0.0127a 1.8020 � 0.0094b Shank circumference (cm) 20.521 � 0.0008a 18.812 � 0.0009b Hoof circumference (cm) 43.013 � 0.0018a 39.751 � 0.0027b a,bMeans in the same row followed by different superscript letters differ significantly at the 5% level (Tukey test). was 0.0364 or 3.64% (0%-7.81%) for the 12 inbred animals of the racing line and 0.0036 (or 0.36%) for all racing animals. These coefficients were higher than those obtained for the only inbred animal of the cutting line (0.0078 or 0.78%) and for all cutting animals (0.0001 or 0.01%), suggesting a higher frequency of inbreeding in the racing line. Coeffi- cients of 0.8% [16] to 1% [17] have been reported for Thor- oughbreds, and of 10.3% for a population of Lipizzan mares [18]. In this research, as well as in the study by McCue et al. [19], the coefficients of inbreeding of Quarter Horses were found to be low when compared with other breeds, indi- cating that Quarter Horse breeders tend to avoid mating between closely related animals. Quality control of individual genotype data led to the exclusion of four cutting animals from the sample because of a call rate <0.95 (95%). The remaining 184 animals (120 racing and 64 cutting animals) presented a mean call rate of 0.9929 � 0.0054 (range: 0.9503- 0.9979). Mean heterozygosity estimated for all individ- uals was 0.3468 � 0.0148 (range: 0.3049-0.3788). The samples genotyped in duplicate showed �99.8% agree- ment (0.9979-0.9990). Similarly, parentage concordance between stallion/progeny and stallion/mare/progeny pairs was very high (0.9985-0.9995). Although some individuals were excluded because of a low genotyping rate, taken together, the present results indicate the lack of DNA contamination between samples and a generally successful hybridization between DNA and the chip. Considering the whole population (n ¼ 184) that passed quality control, 12,544 SNPs were excluded by the filtering process and 42,058 remained for further analysis. With respect to the 120 racing and 64 cutting animals, after filtering, 40,787 (13,815 excluded) and 41,232 (13,370 excluded) SNPs remained for analysis, respectively. Table 2 shows the number of excluded SNPs and the reason for exclusion for the whole population and for animals of the racing and cutting lines. According to Wiggans et al. [20] and Ziegler [21], inconsistently genotyped SNPs or those that do not contribute to the accuracy of genetic evaluations should be excluded to reduce computational effort and the number of false results, as well as to improve precision of the esti- mates of the remaining polymorphisms. The numbers of excluded SNPs owing to an MAF <0.05 were similar in the racing line (8,925), cutting line (7,958), and whole Table 2 Number of SNPs excluded by quality control of genotype data considering the whole population, racing line, and cutting line of Quarter Horses and reason for exclusion Category/Reason for Exclusion Whole Population n Racing Line n Cutting Line n Genotyped SNPs 54,602 54,602 54,602 Located on the X chromosome 2,539 2,539 2,539 Cluster separation <0.3 1,972 1,972 1,972 Call frequency <0.9 148 143 227 Fixed or with minor allele frequency <0.05 7,587 8,925 7,958 HardyeWeinberg equilibriumdP < .001 298 236 674 SNPs selected for genetic analysis 42,058 40,787 41,232 C.T. Meira et al. / Journal of Equine Veterinary Science 33 (2013) 244-249 247 population (7,587). Schröder et al. [22] reported a similar number in a study of Quantitative Trait Loci (QTL) for show jumping in Hanoverian Warmblood horses (7,875 excluded SNPs). In contrast, Lykkjen et al. [23] excluded 13,265 SNPs with an MAF <0.05 in a study identifying QTLs associated with osteochondrosis in Norwegian Standardbred trotters. In the present study, the number of informative SNPs in the two lines (racing animals: 40,787; cutting animals: 41,232) and in the whole population (42,058) were lower than the 47,699 SNPs found in Quarter Horses by McCue et al. [19]. This divergence of results is probably due to different criteria adopted in the filtering process of the genotyped SNPs used in their research, particularly the exclusion of polymorphismswith anMAF<0.01. In contrast, the present results were similar to those reported in studies that investigated other horse breeds using the Equine SNP50 BeadChip, such as Thoroughbreds (40,977) [24], Standard- breds (41,170) [23], Hanoverian horses (43,441) [22], and French trotters (41,249) [25]. Despite the difference in the number of informative SNPs between the racing and cutting lines, autosomal coverage, although higher in the latter, did not vary considerably. Comparison of the mean MAF of the genotyped SNPs revealed no significant difference between the racing (0.2267� 0.1543) and cutting lines (0.2256� 0.1496; F test, P > .05). Similar values have been reported by Corbin et al. [26] and McCue et al. [19], who studied linkage disequi- librium (LD) in Thoroughbred (0.30 � 0.12) and Quarter horses (0.232), respectively. However, the frequency of polymorphisms with an MAF <0.05 was higher in racing horses (17.87%) than in cutting horses (15.96%). Among these, the frequency of fixed (monomorphic) poly- morphisms and those with rare alleles were 6.79% and 11.08% in racing animals and 5.11% and 10.85% in cutting animals, respectively. In addition, the number of informa- tive SNPs (MAF �0.05) was higher in the cutting line (84.04%) than in the racing line (82.13%). Table 3 shows the number of SNPs according to MAF class. The higher frequency of polymorphisms with an MAF <0.05 in racing Quarter Horses might be explained by the more common practice of consanguineous mating and higher selection intensity in this line. Polymorphic SNPs that are associated with one population but not with the other, which were more frequent in the cutting line, might be used as infor- mative SNPs for a line in particular. In contrast, highly informative SNPs (MAF ¼ 0.4-0.5), which were more frequent in the racing line, are useful for individual iden- tification by DNA testing (paternity and forensic tests). Table 4 shows the distribution and density of informa- tive SNPs (MAF �0.05) across the 31 autosomes in the sample of Quarter Horses studied. Considering the whole population, the largest number of informative SNPs was Table 3 Number and percentage of SNPs divided into four classes of minor allele frequen whole population Line Minor Allele Frequency <0.05 0.05-0.5 Racing 8,925 (17.87%) 41,023 (82 Cutting 7,958 (15.96%) 41,906 (84 Whole population 7,587 (15.19%) 42,356 (84 detected on chromosome 1 (3,502), and the smallest number, on chromosome 31 (492). With respect to genome coverage, the density of SNPs was 18.72 SNPs/million base pairs (Mb) or one SNP at an interval of 53,419 bp. This density varied between chromosomes, ranging from 16.82 SNPs/Mb (ECA12: average of one SNP at an interval of 59,453 bp) to 19.91 SNPs/Mb (ECA16: average of one SNP at an interval of 52,356 bp). In the racing line, chromosome 1 contained the largest (3,381) and chromosome 31 con- tained the smallest number of informative SNPs (486). The density was 18.15 SNPs/Mb or one SNP at an interval of 55,096 bp, ranging from 16.09 SNPs/Mb (ECA12: average of one SNP at an interval of 62,150 bp) to 19.44 SNPs/Mb (ECA31: average of one SNP at an interval of 51,440 bp). In the cutting line, ECA1 contained the largest (3,433) and ECA31 contained the smallest number of informative SNPs (496). The lowest SNP density was observed for chromo- some 12, with 16.67 SNPs/Mb (average of one SNP at an interval of 59,988 bp), and the highest density was observed for chromosome 31, with 19.84 SNPs/Mb (average of one SNP at an interval of 50,403 bp), corresponding to a mean interval of 54,496 bp between SNPs distributed across the entire autosomal genome or to a density of 18.35 SNPs/Mb. Equine chromosomes differ significantly in length, with the longest chromosome (ECA1) containing 186Mb and the shortest chromosome (ECA31) containing 25 Mb [27]. Therefore, the higher frequency of informative poly- morphisms on larger chromosomes and the lower frequency on smaller chromosomes observed in the two lines and in the whole Quarter Horse population indicate that the filtering process did not affect the uniformity of distribution of SNPs across the genome. The FST mean estimated for the 42,058 SNPs in the two lines was 0.0342 � 0.0403, with w ranging from 0.0025 to 0.3556. This value was substantially lower than that re- ported by Gu et al. [28], who used microsatellite markers to identify genome regions that distinguish Thoroughbreds from non-Thoroughbred breeds and found an overall mean FST of 0.12. The low FST value found in the present study was expected, as fixation indices tend to be lower when lines instead of breeds are studied because of a theoretically lower genetic distance. Qanbari et al. [29], using FST to quantify genetic differentiation in Bos taurus cattle, found lower values within milk-producing (Holstein and Brown Swiss; 0.057 � 0.076) and meat-producing breeds (Angus and Piedmontese; 0.022 � 0.041) than between these breeds (mean of 0.278 � 0.324). The distribution of posterior means of the w values (FST) in the two lines permitted classification of the 42,058 SNPs (loci) into seven clusters. According to Gianola et al. [15], the expectation is that these clusters are representative of cy obtained for the racing and cutting lines of Quarter Horses and for the 0.05-0.4 0.4-0.5 .13%) 31,580 (63.22%) 9,443 (18.91%) .04%) 33,084 (66.35%) 8,822 (17.69%) .81%) 32,985 (66.05%) 9,371 (18.76%) Table 4 Distribution and density of informative SNPs (minor allele frequency �0.05) across the autosomes of racing and cutting Quarter Horses and of the whole population Chromosomee Chromosome Size (Mb) Informative SNPs Density Racing Line Cutting Line Whole Population Racing Line (SNPs/Mb) Cutting Line (SNP/Mb) Whole Population (SNP/Mb) 1e186 3,381 3,433 3,502 18.18 18.46 18.83 2e121 2,238 2,197 2,289 18.50 18.16 18.92 3e119 2,213 2,213 2,261 18.60 18.60 19.00 4e109 1,980 2,053 2,059 18.17 18.83 18.89 5e100 1,773 1,794 1,826 17.73 17.94 18.26 6e85 1,474 1,451 1,510 17.34 17.07 17.76 7e99 1,792 1,782 1,834 18.10 18.00 18.53 8e94 1,720 1,722 1,767 18.30 18.32 18.80 9e84 1,529 1,555 1,578 18.20 18.51 18.79 10e84 1,478 1,484 1,527 17.60 17.67 18.18 11e61 1,102 1,128 1,134 18.07 18.49 18.59 12e33 531 550 555 16.09 16.67 16.82 13e43 739 754 776 17.19 17.53 18.05 14e94 1,703 1,777 1,764 18.12 18.90 18.77 15e92 1,759 1,703 1,767 19.12 18.51 19.21 16e87 1,675 1,676 1,732 19.25 19.26 19.91 17e81 1,492 1,511 1,546 18.42 18.65 19.09 18e83 1,405 1,462 1,455 16.93 17.61 17.53 19e60 1,151 1,144 1,166 19.18 19.07 19.43 20e64 1,184 1,208 1,225 18.50 18.88 19.14 21e58 1,044 1,052 1,087 18.00 18.14 18.74 22e50 904 969 953 18.08 19.38 19.06 23e56 975 986 1,016 17.41 17.61 18.14 24e47 885 882 908 18.83 18.77 19.32 25e40 703 740 738 17.58 18.50 18.45 26e42 765 765 792 18.21 18.21 18.86 27e40 676 719 721 16.90 17.98 18.03 28e46 860 847 885 18.70 18.41 19.24 29e34 599 616 614 17.62 18.12 18.06 30e30 571 563 579 19.03 18.77 19.30 31e25 486 496 492 19.44 19.84 19.68 C.T. Meira et al. / Journal of Equine Veterinary Science 33 (2013) 244-249248 different processes that occur in the populations, such as balancing, directional selection, and neutrality. Cluster 4, which consisted of 2,558 loci, presented the highest values of w. The conditional probability of membership to this cluster was one for 271 loci. Fifteen of these loci, mapped to chromosomes 1, 2, 4, 5, 8, 9, 10, 16, 21, and 30, presented the highest values of w (0.3013-0.3556), indicating genome regions that were more likely to have been subjected to divergent selection between lines. Recently, genomic studies in the Quarter Horse got prominent. In the first sequencing of a horse genome by next-generation sequencing and the first genomic sequence of an individual Quarter Horse mare, Doan et al. [30] identified 3.1 milion SNPs, 193,000 insertion/deletion polymorphisms (InDels), and 282 copy-number variants in relation to the reference Thoroughbred genome. Addi- tionally, these researches genotyped this Quarter Horse for gene mutations of known diseases and for gene variants associated with economically important traits, including racing performance. The four genes related to racing performance in Thoroughbred horses, MSTN (myostatin) [24,31-33], PDK4 (piruvate dehydrogenase kinase, isozyme 4) [28,34], CKM (creatine kinase, muscle), and COX4I2 (cytochrome c oxidase, subunit 4, isoform 2) [35], are located in equine chromosomes 18, 4, 10, and 22, respec- tively. These chromosomes are shown in this research as having 77 of 271 genomic regions (with conditional prob- ability equal to one) selected divergently between racing and cutting lines. However, considering that the LD decrease in the Quarter Horse occurs between loci that are located at distances longer than 50-100 kb [19], none of the 77 regions close enough (in LD) to the gene polymorphisms associated with racing performance. Considering that speed is also a desirable trait in cutting animals, these results suggest the possibility that the same alleles of these gene polymorphisms have been selected in both lines of Quarter Horse. 4. Conclusions Divergent selection of cutting and racing Quarter Horses promoted significant changes in the physical and genomic characteristics of these lines. The number of informative SNPs and SNP density found in the genome of cutting and racing Quarter Horses suggest that the Equine SNP50 BeadChip can be used for different purposes in the breed, such as genetic structure analysis, estimation of genetic divergence within and between populations, and identifi- cation of QTL and genome regions subjected to selection. Acknowledgments The authors thank the following horse farms for providing the Quarter Horses: Haras Fazenda Bela, Capela do Alto/SP; Haras Vista Verde, Boituva/SP; Central de Reprodução Rancho das Américas, Porto Feliz/SP; Haras C.T. Meira et al. / Journal of Equine Veterinary Science 33 (2013) 244-249 249 Carrera, Avaré/SP; Haras São Jorge, Tatuí/SP; Xandi Ranch, Bauru/SP; Haras Villa San José, Porto Feliz/SP; and Haras Rancho Emanuel, Porto Feliz/SP. The present research was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo and fellowships were granted by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior. References [1] ABQM. Associação brasileira dos criadores de cavalos Quarto de Milha. Available at: http://www.abqm.com.br. Accessed December 12, 2011. [2] Kijas JW, Townley D, Dalrymple BP, Heaton MP, Maddox JF, McGrath A, et al. A genome wide survey of SNP variation reveals the genetic structure of sheep breeds. PLoS One 2009;4:1-13. [3] McKay SD, Schnabel RD, Murdoch BM, Matukumalli LK, Aerts J, Coppieters W, et al. An assessment of population structure in eight breeds of cattle using a whole genome SNP panel. BMC Genet 2008;9:1-9. [4] Gibbs RA, Taylor JF, Van Tassell CP. Genome-wide survey of SNP variation uncovers the genetic structure of cattle breeds. Science 2009;324:528-32. [5] Zenger KR, Khatkar MS, Cavanagh JA, Hawken RJ, Raadsma HW. Genome-wide genetic diversity of Holstein Friesian cattle reveals new insights into Australian and global population variability, including impact of selection. Anim Genet 2006;38:7-14. [6] Andersson L, Georges M. Domestic-animal genomics: deciphering the genetics of complex traits. Nat Rev Genet 2004;5:202-12. [7] Chowdhary BP, Raudsepp T. The Horse Derby: racing from map to whole genome sequence. Chromosome Res 2008;16:109-27. [8] Wade CM, Giulotto E, Sigurdsson S, Zoli M, Gnerre S, Imsland F, et al. Genome sequence, comparative analysis, and population genetics of the domestic horse. Science 2009;326:865-7. [9] Torres AP, Jardim WR. Criação do cavalo e de outros eqüinos. São Paulo, Brazil: Livraria Nobel; 1992. [10] SAS. Statistical Analysis System, 2004 SAS/STAT user’s guide, version 9.1, Vol 1-7. Cary, NC: SAS Institute Inc. [11] Strandén I, Vuori K. Relax2: pedigree analysis program. In: Proceed- ings of the 8th World Congress on Genetics Applied to Livestock Production; August 13-18, 2006; Belo Horizonte, MG, Brasil. [12] Wright S. The genetical structure of populations. Ann Eugen 1951;15:323-54. [13] Cockerham CC. Variance of gene frequencies. Evolution 1969;23: 72-84. [14] Weir BS, Hill WG. Estimating F-statistics. Annu Rev Genet 2002;36: 721-50. [15] Gianola D, Simianer H, Qanbari S. A two-step method for detecting selection signatures using genetic markers. Genet Res 2010;92: 141-55. [16] Abrahão AR, Mota MDS, Oliveira HN, Madureira AP. Endogamia em éguas da raça Puro-Sangue Inglês. In: Anais IV Simpósio Nacional de Melhoramento Animal; 2002; Campo Grande, Brasil. [17] Taveira RZ, Mota MDS, Oliveira HN, Abrahão AR, Madureira AP. Endogamia em cavalos da raça Puro-Sangue Inglês. In: Proceedings of the 47� Congresso Nacional de Genética; October 2-5, 2001; Águas de Lindóia, SP, Brasil. [18] Curik I, Zechner P, Sölkner J, Achmann R, Bodo I, Dovc P, et al. Inbreeding, microsatellite heterozygosity, and morphological traits in Lipizzan horses. J Hered 2003;94:125-32. [19] 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;8:1-14. [20] Wiggans GR, Sonstegard TS, VanRaden PM, Matukumalli LK, Schnabel RD, Taylor JF, et al. Selection of single-nucleotide poly- morphisms and quality of genotypes used in genomic evaluation of dairy cattle in the United States and Canada. J Dairy Sci 2009;92:3431-6. [21] Ziegler A. Genome-wide association studies: quality control and population-based measures. Genet Epidemiol 2009;33(Suppl 1): S45-50. [22] Schröder W, Klostermann A, Stock KF, Distl O. A genome-wide association study for quantitative trait loci of show-jumping in Hanoverian Warmblood horses. Anim Genet 2011;43:392-400. [23] Lykkjen S, Dolvik NI, McCue ME, Rendahl AK, Mickelson JR, Roed KH. Genome-wide association analysis of osteochondrosis of the tibio- tarsal joint in Norwegian Standardbred trotters. Anim Genet 2010;41:111-20. [24] Hill EW, McGivney BA, Gu J, Whiston R, MacHugh DE. A genome- wide SNP-association study confirms a sequence variant (g.66493737C>T) in the equine myostatin (MSTN) gene as the most powerful predictor of optimum racing distance for Thoroughbred racehorses. BMC Genomics 2010;11:1-10. [25] Teyssèdre S, Dupuis MC, Guérin G, Schibler L, Denoix JM, Elsen JM, et al. Genome-wide association studies for osteochondrosis in French Trotters. J Anim Sci 2012;90:45-53. [26] Corbin LJ, Blott SC, Swinburne JE, Vaudin M, Bishop SC, Woolliams JA. Linkage disequilibrium and historical effective pop- ulation size in the Thoroughbred horse. Anim Genet 2010;41:8-15. [27] NCBI. National Center for Biotechnology Information. Available at: http://www.ncbi.nlm.nih.gov. Accessed November 7, 2011. [28] Gu J, Orr N, Park SD, Katz LM, Sulimova G, MacHugh DE, et al. A genome scan for positive selection in Thoroughbred horses. PLoS One 2009;4:1-17. [29] Qanbari S, Gianola D, Hayes B, Schenkel F, Miller S, Moore S, et al. Application of site and haplotype-frequency based approaches for detecting selection signatures in cattle. BMC Genomics 2011;12:1- 12. [30] Doan R, Cohen ND, Sawyer J, Ghaffari N, Johnson CD, Dindot SV. Whole-genome sequencing and genetic variant analysis of a Quarter Horse mare. BMC Genomics 2012;13:1-11. [31] Hill EM, Gu J, Eivers SS, Fonseca RG, McGiveny BA, Govindarajan P, et al. A sequence polymorphism in MSTN predicts sprinting ability and racing stamina in Thoroughbred horses. PLoS One 2010;5:1-6. [32] Binns MM, Boehler DA, Lambert DH. Identification of the myo- statin locus (MSTN) as having a major effect on optimum racing distance in the Thoroughbred horse in the USA. Anim Genet 2010;41:154-8. [33] Tozaki T, Miyake T, Kakoi H, Gawahara H, Sugita S, Hasegawa T, et al. A genome-wide association study for racing performances in Thoroughbreds clarifies a candidate region near the MSTN gene. Anim Genet 2010;41:28-35. [34] Hill EM, Gu J, McGiveny BA, MacHugh DE. Targets of selection in the Thoroughbred genome contain exercise-relevant gene SNPs associated with elite racecourse performance. Anim Genet 2010; 41:56-63. [35] Gu J, MacHugh DE, McGiveny BA, Park SDE, Katz LM, Hill EM. Association of sequence variants in CKM (creatine kinase, muscle) and COX4I2 (cytochrome c oxidase, subunit 4, isoform 2) genes with racing performance in Thoroughbred horses. Equine Vet J 2010;42: 569-75. http://www.abqm.com.br http://www.ncbi.nlm.nih.gov Morphological and Genomic Differences Between Cutting and Racing Lines of Quarter Horses 1. Introduction 2. Materials and Methods 2.1. Animals and Phenotypic Data 2.2. Genotyping of SNPs 2.3. Analysis of Morphological Differences and Differences in Inbreeding between Cutting and Racing Lines 2.4. Quality of Genotype Data 2.5. Analysis of Genomic Differences between Cutting and Racing Lines 3. Results and Discussion 4. Conclusions Acknowledgments References