Detection of selection signatures in dairy and beef cattle using high-density genomic information

Background

Artificial selection for economically important traits in cattle is expected to have left distinctive selection signatures on the genome. Access to high-density genotypes facilitates the accurate identification of genomic regions that have undergone positive selection. These findings help to better elucidate the mechanisms of selection and to identify candidate genes of interest to breeding programs.

Results

Information on 705 243 autosomal single nucleotide polymorphisms (SNPs) in 3122 dairy and beef male animals from seven cattle breeds (Angus, Belgian Blue, Charolais, Hereford, Holstein-Friesian, Limousin and Simmental) were used to detect selection signatures by applying two complementary methods, integrated haplotype score (iHS) and global fixation index (F_ST). To control for false positive results, we used false discovery rate (FDR) adjustment to calculate adjusted iHS within each breed and the genome-wide significance level was about 0.003. Using the iHS method, 83, 92, 91, 101, 85, 101 and 86 significant genomic regions were detected for Angus, Belgian Blue, Charolais, Hereford, Holstein-Friesian, Limousin and Simmental cattle, respectively. None of these regions was common to all seven breeds. Using the F_ST approach, 704 individual SNPs were detected across breeds. Annotation of the regions of the genome that showed selection signatures revealed several interesting candidate genes i.e. DGAT1, ABCG2, MSTN, CAPN3, FABP3, CHCHD7, PLAG1, JAZF1, PRKG2, ACTC1, TBC1D1, GHR, BMP2, TSG1, LYN, KIT and MC1R that play a role in milk production, reproduction, body size, muscle formation or coat color. Fifty-seven common candidate genes were found by both the iHS and global F_ST methods across the seven breeds. Moreover, many novel genomic regions and genes were detected within the regions that showed selection signatures; for some candidate genes, signatures of positive selection exist in the human genome. Multilevel bioinformatic analyses of the detected candidate genes suggested that the PPAR pathway may have been subjected to positive selection.

Conclusions

This study provides a high-resolution bovine genomic map of positive selection signatures that are either specific to one breed or common to a subset of the seven breeds analyzed. Our results will contribute to the detection of functional candidate genes that have undergone positive selection in future studies.

Electronic supplementary material

The online version of this article (doi:10.1186/s12711-015-0127-3) contains supplementary material, which is available to authorized users.     

Genetic parameters for carcass weight,conformation and fat in five beef cattle breeds

Profitability of beef production can be increased by genetically improving carcass traits. To construct breeding value evaluations for carcass traits, breed-specific genetic parameters were estimated for carcass weight, carcass conformation and carcass fat in five beef cattle breeds in Finland (Hereford, Aberdeen Angus, Simmental, Charolais and Limousin). Conformation and fat were visually scored using the EUROP carcass classification. Each breed was separately analyzed using a multitrait animal model. A total of 6879–19 539 animals per breed had phenotypes. For the five breeds, heritabilities were moderate for carcass weight (h²=0.39 to 0.48, s.e.=0.02 to 0.04) and slightly lower for conformation (h²=0.30 to 0.44, s.e.=0.02 to 0.04) and carcass fat (h²=0.29 to 0.44, s.e.=0.02 to 0.04). The genetic correlation between carcass weight and conformation was favorable in all breeds (r_G=0.37 to 0.53, s.e.=0.04 to 0.05), heavy carcasses being genetically more conformed. The phenotypic correlation between carcass weight and carcass fat was moderately positive in all breeds (r_P=0.21 to 0.32), implying that increasing carcass weight was related to increasing fat levels. The respective genetic correlation was the strongest in Hereford (r_G=0.28, s.e.=0.05) and Angus (r_G=0.15, s.e.=0.05), the two small body-sized British breeds with the lowest conformation and the highest fat level. The correlation was weaker in the other breeds (r_G=0.08 to 0.14). For Hereford, Angus and Simmental, more conformed carcasses were phenotypically fatter (r_P=0.11 to 0.15), but the respective genetic correlations were close to zero (r_G=−0.05 to 0.04). In contrast, in the two large body-sized and muscular French breeds, the genetic correlation between conformation and fat was negative and the phenotypic correlation was close to zero or negative (Charolais: r_G=−0.18, s.e.=0.06, r_P=0.02; Limousin: r_G=−0.56, s.e.=0.04, r_P=−0.13). The results indicate genetic variation for the genetic improvement of the carcass traits, favorable correlations for the simultaneous improvement of carcass weight and conformation in all breeds, and breed differences in the correlations of carcass fat.     

Breed variation and genetic parameters for growth and body development in diverse beef cattle genotypes

Conformation scores can account for more than 20% of cattle price variation at Australian livestock sales. However, there are limited available references which define genetic factors relating objective live developmental traits to carcass composition. Weaning and post-weaning weight, height, length, girth, muscle (ratio of stifle to hip width) and fat depth of 1202 progeny from mature Hereford cows (637) mated to seven sire breeds (Jersey, Wagyu, Angus, Hereford, South Devon, Limousin and Belgian Blue) were examined for growth and development across ages. Crossbred Wagyu and Jersey were both lighter in weight and smaller in size (height, length and girth) than purebred Hereford and crossbred Angus, South Devon, Limousin and Belgian Blue. Within the five larger crossbreds, there were significant changes in relative weight from weaning to 600 days. Sire breeds differed in fat depth, with Angus being the fattest (9% on average fatter than Hereford and Wagyu), and Jersey 5% less fat than Hereford, followed by South Devon and Limousin (19% lower than Hereford) and Belgian Blue (39% lower than Hereford). Direct heritability ranged from 19 to 42% and was higher than the proportion of total phenotypic variance accounted for by maternal effects (which ranged from 0 to 17%) for most body measurement traits except for weight (38 v. 18%) and girth (36 v. 9%) traits at weaning, an indication of maternal effect on some body conformation traits at early ages. Muscularity (19 to 44%) and fat depth (26 to 43%) were moderately to highly heritable across ages. There were large differences for growth and the objective measures of body development between crossbreds with a degree of overlap among the progeny of the seven sire breeds. The variation between genetic (positive) and environmental (negative) correlations for dry versus wet season average daily gains in weight and fat, suggested the potential use of live-animal conformation traits for within breed selection of genetically superior animal in these traits across seasons.     

Scrotal circumference of two-year-old bulls of several beef breeds

Scrotal circumference (SC) was measured on 7,918 2-yr-old Angus, Charolais, horned and polled Herefords, Limousin, Shorthorn, and Simmental bulls presented to culling committees at six show/sales between 1977 and 1983. Only SC data from bulls within the age range of 24 +/- 4 mo were used. Scrotal circumference data were corrected across breeds for the effects of location-year and sire and were adjusted to a common bull age of 730 d. The adjusted mean SC (+/- SE) for 2-yr-old beef bulls was Simmental, 38.8 +/- 0.10 cm (n = 540); Aberdeen Angus, 37.2 +/- 0.09 cm (n = 629); Charolais, 36.3 +/- 0.09 cm (n = 499); horned Hereford, 36.1 +/- 0.03 cm (n = 3,769); polled Hereford, 35.6 +/- 0.04 cm (n = 2,170); Shorthorn, 34.9 +/- 0.11 cm (n = 231); and Limousin, 32.2 +/- 0.18 cm (n = 80). The authors' recommendations of minimum acceptable SC for 2-yr-old beef bulls are Simmental, 36.0 cm; Angus and Charolais, 35.0 cm; horned and polled Herefords and Shorthorn, 34.0 cm; and Limousin, 33.0 cm.     

Real-time PCR genotyping and frequency of the myostatin F94L mutation in beef cattle breeds

This research developed two real-time PCR assays, employing high-resolution melt and allele-specific analysis to accurately genotype the F94L mutation in cattle. This mutation (g.433C > A) in the growth differentiation factor 8 or myostatin gene has recently been shown to be functionally associated with increased muscle mass and carcass yield in cattle. The F94L mutation is not, like other myostatin mutations, associated with reduced fertility and dystocia. It is therefore a candidate for introgression into other breeds to improve retail beef yield and the development of a simple and accurate test to genotype this specific mutation is warranted. Variations in the efficiency of enzyme cleavage compromised the accuracy of genotyping by published methods, potentially resulting in an overestimation of the frequency of the mutant allele. The frequency of the F94L mutation was determined by real-time PCR in 1140 animals from 15 breeds of cattle in Australia. The mutation was present in Simmental (0.8%), Piedmontese (2%), Droughtmaster (4%) and Limousin (94.2%) but not found in Salers, Angus, Poll Hereford, Hereford, Gelbvieh, Charolais, Jersey, Brahman, Holstein, Shorthorn or Maine Anjou. The low prevalence of F94L in all beef breeds except Limousin indicates the significant potential for this mutation to improve retail yield in Australian beef cattle.     

Genetic relationships between feral cattle from Chirikof Island, Alaska and other breeds

The origin of cattle on Chirikof Island, off the coast of Alaska, is not well documented. We assessed genetic differentiation of cattle isolated on Chirikof Island from several breeds commonly used for commercial production in North America including breeds popularly believed to have contributed to the Chirikof Island population. A set of 34 microsatellite loci was used to genotype Angus, Charolais, Hereford, Highland, Limousin, Red Angus, Salers, Shorthorn, Simmental, Tarentaise and Texas Longhorn cattle sampled from North America and the Chirikof Island population. Resulting F(ST) statistics for these loci ranged from 0.06 to 0.22 and on average, 14% of total genetic variation was between breeds. Whether population structure was modelled as a bifurcating tree or genetic network, Chirikof Island cattle appeared to be unique and strongly differentiated relative to the other breeds that were sampled. Bayesian clustering for multiple-locus assignment to genetic groups indicated low levels of admixture in the Chirikof Island population. Thus, the Chirikof Island population may be a novel genetic resource of some importance for conservation and industry.     

肉牛杂交优势预测、评估及其应用研究

利用微卫星标记技术分析了８个肉 要交样本群体的遗传结构和遗传变异,预测了要种优势。在此基础上肉牛发校组合的实际杂交效果利用个体动物模型进行了评估,并提出了筛选最优杂交组合的分子数量遗传学综合评选新方法。其最优杂交组合的评选结果为,在丰宁、隆化代表区域,以海伏特、利木赞和夏洛来为父本的组合最好;在赞皇代表区域,以利木赞、安格斯和海伏特为父本的组合最好,在抚宁代表区域,以海伏特、利木赞和皮埃蒙特为父本     

Breed differences in the distribution of BoLA-A locus antigens in American cattle

A total of 627 cattle representing seven breeds from south central Nebraska, USA were tested for 37 BoLA antigens which behave as products of 37 distinct alleles of the class I BoLA-A locus. Four antigens were absent from all breeds tested. The other antigens showed marked and statistically significant differences in breed distribution. There was no evidence for blank (null) alleles. The number of alleles in each breed ranged from 10 to 20. The Hereford and Simmental populations tested were less polymorphic than the Angus, Brown Swiss, Charolais, Gelbvieh and Limousin populations.     

Nucleotide Substitution in 3' Arm of Bovine MIR-2467 in Five Cattle Breeds

The T > C single nucleotide polymorphism (SNP) in the MIR2467 gene was investigated in order to confirm its presence in cattle genome and to check for possible differences in its genotype distribution among different breeds. Additional purpose of the study was to investigate in silico potential effect of that substitution on the structure and stability of precursor mir-2467. The study involved 634 individuals of five cattle breeds: Angus, Hereford, Holstein-Friesian, Jersey, and Limousin, which were genotyped using PCR-RFLP assay. In this study, the presence of T > C polymorphism at position 24 was observed in all the cattle breeds excepting Hereford. In addition, the differences in the genotype distribution among analyzed breeds were indicated. On the basis of minimum free energy structure prediction, the C allele was indicated to have possible impact on decreasing the stability of the pre-mir-2467, thus altering its ability to regulate target genes expression.     

A potential association between the BM 1500 microsatellite and fat deposition in beef cattle

The obese gene was hypothesized as a candidate gene for fat characteristics in beef cattle. The BM 1500 microsatellite, near the obese gene, was characterized in 158 purebred beef bulls for which carcass trait information was available. Four breeds were included in the analyses—Angus, Charolais, Hereford, and Simmental. Four alleles were found. Lengths were approximately 138, 147, 149, and 140 bp with genotypic frequencies of 0.47, 0.44, 0.09, and 0.003 respectively. The carcass traits %rib fat, %rib lean, average fat, and grade fat were found to be significantly associated with the different alleles. The presence of the 138-bp allele in the genotype of an animal is correlated with higher levels of fat, whereas the 147-bp allele has the opposite effect. The 149-bp allele was found in low numbers, and a homozygote was never identified. Hereford and Angus bulls had the greatest frequencies of 138-bp alleles (Hereford = 0.57, Angus = 0.59), while Charolais and Simmental had a greater proportion of 147-bp alleles (Charolais = 0.54, Simmental = 0.58). This information may aid cattle producers in selecting cattle for markets that differ in the amount of fat required. Received: 27 October 1997 / Accepted: 23 January 1998     

Joint analysis of beef growth and carcass quality traits through calculation of co-variance components and correlations

A joint growth-carcass model using random regression was used to estimate the (co)variance components of beef cattle body weights and carcass quality traits and correlations between them. During a four-year period (1994-1997) of the Australian "southern crossbreeding project", mature Hereford cows (N = 581) were mated to 97 sires of Jersey, Wagyu, Angus, Hereford, South Devon, Limousin, and Belgian Blue breeds, resulting in 1141 calves. Data included 13 (for steers) and 8 (for heifers) body weight measurements approximately every 50 days from birth until slaughter and four carcass quality traits: hot standard carcass weight, rump fat depth, rib eye muscle area, and intramuscular fat content. The mixed model included fixed effects of sex, sire breed, age (linear, quadratic and cubic), and their interactions between sex and sire breed with age. Random effects were sire, dam, management (birth location, year, post-weaning groups), and permanent environmental effects, and their interactions with linear, quadratic and cubic growth, when possible. Phenotypic, sire and dam correlations between body weights and hot standard carcass weight and rib eye muscle area were positive and moderate to high from birth to feedlot period. Management variation accounted for the largest proportion of total variation in both growth and carcass traits. Management correlations between carcass traits were high, except between rump fat depth and intramuscular fat (r = 0.26). Management correlations between body weight and carcass traits during the pre-weaning period were positive except for intramuscular fat. The correlations were low from birth to weaning, then increased dramatically and were high during the feedlot period.     

Comparison of molecular breeding values based on within- and across-breed training in beef cattle

Background

Although the efficacy of genomic predictors based on within-breed training looks promising, it is necessary to develop and evaluate across-breed predictors for the technology to be fully applied in the beef industry. The efficacies of genomic predictors trained in one breed and utilized to predict genetic merit in differing breeds based on simulation studies have been reported, as have the efficacies of predictors trained using data from multiple breeds to predict the genetic merit of purebreds. However, comparable studies using beef cattle field data have not been reported.

Methods

Molecular breeding values for weaning and yearling weight were derived and evaluated using a database containing BovineSNP50 genotypes for 7294 animals from 13 breeds in the training set and 2277 animals from seven breeds (Angus, Red Angus, Hereford, Charolais, Gelbvieh, Limousin, and Simmental) in the evaluation set. Six single-breed and four across-breed genomic predictors were trained using pooled data from purebred animals. Molecular breeding values were evaluated using field data, including genotypes for 2227 animals and phenotypic records of animals born in 2008 or later. Accuracies of molecular breeding values were estimated based on the genetic correlation between the molecular breeding value and trait phenotype.

Results

With one exception, the estimated genetic correlations of within-breed molecular breeding values with trait phenotype were greater than 0.28 when evaluated in the breed used for training. Most estimated genetic correlations for the across-breed trained molecular breeding values were moderate (> 0.30). When molecular breeding values were evaluated in breeds that were not in the training set, estimated genetic correlations clustered around zero.

Conclusions

Even for closely related breeds, within- or across-breed trained molecular breeding values have limited prediction accuracy for breeds that were not in the training set. For breeds in the training set, across- and within-breed trained molecular breeding values had similar accuracies. The benefit of adding data from other breeds to a within-breed training population is the ability to produce molecular breeding values that are more robust across breeds and these can be utilized until enough training data has been accumulated to allow for a within-breed training set.     

Modeling of crossbred cattle growth, comparison between cubic and piecewise random regression models

Two analyses, cubic and piecewise random regression, were conducted to model growth of crossbred cattle from birth to about two years of age, investigating the ability of a piecewise procedure to fit growth traits without the complications of the cubic model. During a four-year period (1994-1997) of the Australian "Southern Crossbreeding Project", mature Hereford cows (N = 581) were mated to 97 sires of Angus, Belgian Blue, Hereford, Jersey, Limousin, South Devon, and Wagyu breeds, resulting in 1141 steers and heifers born over four years. Data included 13 (for steers) and eight (for heifers) live body weight measurements, made approximately every 50 days from birth until slaughter. The mixed model included fixed effects of sex, sire breed, age (linear, quadratic and cubic), and their interactions between sex and sire breed with age. Random effects were sire, dam, management (birth location, year, post-weaning groups), and permanent environmental effects and for each of these when possible, their interactions with linear, quadratic and cubic growth. In both models, body weights of all breeds increased over pre-weaning period, held fairly steady (slightly flattening) over the dry season then increased again towards the end of the feedlot period. The number of estimated parameters for the cubic model was 22 while for the piecewise model it was 32. It was concluded that the piecewise model was very similar to the cubic model in the fit to the data; with the piecewise model being marginally better. The piecewise model seems to fit the data better at the end of the growth period.     

SNP-based association mapping of the polled gene in divergent cattle breeds

Naturally, hornless cattle are called polled. Although the POLL locus could be assigned to a c. 1.36‐Mb interval in the centromeric region of BTA1, the underlying genetic basis for the polled trait is still unknown. Here, an association mapping design was set up to refine the candidate region of the polled trait for subsequent high‐throughput sequencing. The case group comprised 101 homozygous polled animals from nine divergent cattle breeds, the majority represented by Galloway, Angus, Fleckvieh and Holstein Friesian. Additionally, this group included some polled individuals of Blonde d’Aquitaine, Charolais, Hereford, Jersey and Limousin breeds. The control group comprised horned Belgian Blue, Fleckvieh, Holstein Friesian and Illyrian Bu?a cattle. A genome‐wide scan using 49 163 SNPs was performed, which revealed one shared homozygous haplotype block consisting of nine neighbouring SNPs in all polled animals. This segment defines a 381‐kb interval on BTA1 that we consider to be the most likely location of the POLL mutation. Our results further demonstrate that the polled‐associated haplotype is also frequent in horned animals included in this study, and thus the haplotype as such cannot be used for population‐wide genetic testing. The actual trait‐associated haplotype may be revealed by using higher‐density SNP arrays. For the final identification of the causal mutation, we suggest high‐throughput sequencing of the entire candidate region, because the identification of functional candidate genes is difficult owing to the lack of a comparable model.     

Sequence comparison of mitochondrial tRNA genes and origin of light strand replication in Bos taurus and Nellore (Bos indicus) breeds

Although the complete bovine mitochondrial DNA molecule has been previously sequencedand sequence comparisons of the mitochondrial displacement loop have been performed, detailed sequence information is limited on coding regions of mitochondrial DNA within and among breeds of Bos taurus and Bos indicus. This study analysed polymorphism of the mitochondrial DNA transfer RNA genes for trypto-phan, alanine, asparagine, cysteine, tyrosine and the origin of light strand replication among Ayrshire, Canadian, Belgium Blue, Brown Swiss, Hereford, Jersey, Limousine, Piedmon-taise, Red Angus, Simmental (Bos taurus) and a Nellore (Bos indicus). Nucleotide sequence analysis of a 420-bp fragment of mitochondrial DNA comprising the five transfer RNA genes showed 100% homology among single individuals of the Bos taurus breeds. The Nellore breed showed guanine to adenine substitutions in the DHU arm of asparagine tRNA and in the origin of light-strand replication. This equates to a 0.5% sequence difference between the Nellore andBos taurus breeds and may reflect an independent evolutionary origin of the species.     

Breed-type effects on fertility traits of young bulls

Measures of fertility in young beef bulls (N=76) representative of divergent breeds and F₁ crosses were evaluated. Breed-types were straightbred Hereford and Red Poll, Hereford × Red Poll reciprocal crosses, Angus × Hereford, Angus × Charolais, Brahman × Hereford and Brahman × Angus.There was evidence of significant breed-type variation in scrotal circumference and total testicle weight, although differences were less pronounced when animal weight was taken into account. Scrotal circumference in Brahman × Hereford bulls was less (P<.05) than straightbred Hereford, Red Poll × Hereford and Angus × Hereford crosses at 8 months of age. However, at 13 months there was no difference in scrotal circumference measurements of Brahman × Hereford, Red Poll × Hereford and Angus × Hereford crosses, and all three crosses exceeded (P<.05) straightbred Herefords. Breed-type differences in semen score and presence of normal, mature sperm in the seminiferous tubules at 13 months of age approached significance at the .10 probability level. Heterosis effects on bull fertility characteristics were insignificant when body weight was included as a covariate in the analyses.     

Prediction model of a joint analysis of beef growth and carcass quality traits

A joint growth-carcass analysis was conducted to develop equations for predicting carcass quality traits associated with variation in growth path of crossbred cattle. During a four-year period (1994-1997) of the Australian "Southern Crossbreeding Project", mature Hereford cows (r = 581) were mated to 97 sires of Jersey, Wagyu, Angus, Hereford, South Devon, Limousin, and Belgian Blue breeds, resulting in 1141 calves. Data included body weight measurements of steers and heifers from birth until slaughter and four carcass quality traits: hot standard carcass weight, rump fat depth, rib eye muscle area, and intramuscular fat content. The model provides nine outputs: median and mean of carcass quality traits, predicted means, and lower and upper confidence intervals, as well as predicted intervals of carcass quality traits (95%) and economic values for domestic market and export markets. Input to the model consists of sex, sire breeds, age (in days)-weight (kg) pairs and slaughter age (500 days for heifer and 700 days for steers). The prediction model is able to accommodate different sexes across seven sire breeds and various management groups at any slaughter age. Its strength lies in its simplicity and flexibility, desirable to accommodate producers with different management schemes. In general, fat depth and intramuscular fat were found to be more affected by differences in growth rate than hot carcass weight and eye muscle area. Also, export market value was more sensitive to growth rate modifications than domestic market value. This model provides a tool by which the producer can estimate the impact of management decisions.     

Analysis of genetic relationships between 10 cattle breeds with 17 microsatellites

To guide genetic conservation programmes with objective criteria, general genetic variability has to be taken into account. This study was conducted to determine the genetic variation between 10 cattle breeds by using 17 microsatellite loci and 13 biochemical markers (11 blood groups, the transferrin and β-casein loci). Microsatellite loci were amplified in 31–50 unrelated individuals from 10 cattle breeds: Charolais, Limousin, Breton Black Pied, Parthenais, Montbéliard, Vosgien, Maine-Anjou, Normande, Jersey and Holstein. Neighbor-joining trees were calculated from genetic distance estimates. The robustness of tree topology was obtained by bootstrap resampling of loci. A total of 210 alleles of the 17 microsatellites were detected in this study and average heterozygosities ranged from 0·53 in the Jersey breed to 0·66 in the Parthenais breed. In general, low bootstrap values were obtained: with the 17 microsatellites, the highest bootstrap values concerned the Holstein/Maine-Anjou grouping with an occurrence of 74%; with the biochemical markers, this node had an occurrence of 79% and the Charolais/Limousin grouping appeared with an occurrence of 74%; when microsatellites and biochemical polymorphism were analysed together, the occurrence of the Holstein/Maine-Anjou grouping was 90% and that of the Charolais/Limousin grouping was 42%. These results suggest that 30 microsatellites, a number currently considered as sufficient to distinguish closely related breeds is, in fact, probably insufficient.     

The use of genetic relationships among cattle breeds in the formulation of rational breeding policies: an example with South Devon (South Africa) and Gelbvieh (Germany)*

At the instigation of the South Devon Cattle Breeders' Society of South Africa we studied the genetic relationships of South Devon cattle (South Africa) and Gelbvieh cattle (Germany) with eight other breeds including Red Angus and Black Angus (USA). On the basis of comparisons at ten immungenetic loci Gelbvieh and South Devon cattle are nearly as genetically similar as Red and Black Angus. The results indicate that Gelbvieh and South Devon had a common ancestry on the Continent and are distinct from other British breeds such as Hereford, Angus and Jersey. This study illustrates the application of studies of the genetic similarities of different breeds to the rational development of future breeding and preservation programs.     

Genome‐wide copy number variation in the bovine genome detected using low coverage sequence of popular beef breeds,

Copy number variations (CNVs) are large insertions, deletions or duplications in the genome that vary between members of a species and are known to affect a wide variety of phenotypic traits. In this study, we identified CNVs in a population of bulls using low coverage next‐generation sequence data. First, in order to determine a suitable strategy for CNV detection in our data, we compared the performance of three distinct CNV detection algorithms on benchmark CNV datasets and concluded that using the multiple sample read depth approach was the best method for identifying CNVs in our sequences. Using this technique, we identified a total of 1341 copy number variable regions (CNVRs) from genome sequences of 154 purebred sires used in Cycle VII of the USMARC Germplasm Evaluation Project. These bulls represented the seven most popular beef breeds in the United States: Hereford, Charolais, Angus, Red Angus, Simmental, Gelbvieh and Limousin. The CNVRs covered 6.7% of the bovine genome and spanned 2465 protein‐coding genes and many known quantitative trait loci (QTL). Genes harbored in the CNVRs were further analyzed to determine their function as well as to find any breed‐specific differences that may shed light on breed differences in adaptation, health and production.