Contents lists available at ScienceDirect Livestock Science journal homepage: www.elsevier.com/locate/livsci Assessment of pedigree information in the Quarter Horse: Population, breeding and genetic diversity Ricardo António Silva Fariaa,⁎, Amanda Merchi Maioranoa, Priscila Arrigucci Bernardesa, Guilherme Luis Pereirab, Marina Gabriela Berchiol Silvab, Rogério Abdallah Curib, Josineudson Augusto II Vasconcelos Silvab a Departmento de Zootecnia, Faculdade de Ciências Agrárias e Veterinárias – Unesp, CEP 14.884-900, Jaboticabal, São Paulo, Brazil bDepartmento Melhoramento e Nutrição Animal, Faculdade de Medicina Veterinária e Zootecnia - Unesp, Botucatu, São Paulo CEP 18.618-307, Brazil A R T I C L E I N F O Keywords: Ancestors Effective population size Equine Inbreeding Population structure A B S T R A C T This study aimed to evaluate population parameters and to describe the genetic diversity of Quarter Horse breed (QH) in Brazil, reported for the first time in the literature. The pedigree data comprised 131,716 animals re- presenting the total population (TP), with records of animals born between 1747 and 2008. The reference population (RP) representing the last generation was applied in this study considering 47,861 animals born between 2000 and 2008. The average generation interval was 9.6 and 10.8 years in TP and RP, respectively. The average equivalent complete generations (EG) were 5.09 (TP) and 6.24 (RP). The inbreeding coefficient (F), average relatedness (AR) and the increase in inbreeding by generation (ΔF) was 1.07%, 0.95% and 0.24%, respectively, for TP. The effective population size (Ne) based on ΔF was 195 and 164 for TP and RP, respectively. The effective number of founders (fe) was 1045 and 811 for TP and RP, respectively, that of ancestors (fa) was 156 and 113, and that of founder genomes (fg) was 105 and 66. The fe/fa and fe/fg ratios in TP were 6.70 and 9.95, respectively, and an increase was observed in RP, indicated a strong bottleneck effect. The total genetic diversity of the QH breed was explained by 4780 ancestors, with 50% of diversity being explained by only 121 and 72 ancestors in TP and RP, respectively. The thoroughbred stallion Three Bars is the most influential an- cestor with the largest marginal genetic contribution for TP (5.73) and RP (5.94%). The results demonstrate a large number of founders and ancestors, but a small ancestor group was responsible for the continuity of the QH breed in Brazil. These finding highlight the importance of monitoring genetic diversity, including follow-up by breeding programs, to permit control of the next generations. 1. Introduction The Quarter Horse (QH) breed originated in the United States in the 17th century from the crossing of stallions from Arabia and Turkey, characterized by resistance and elegance, with fast and muscular English Thoroughbred dams, resulting in a compact and muscular horse that is agile when used as a working horse and fast in short-distance races (ABQM, 2017). The versatility of the QH breed was investigated recently in the genomic study of Petersen et al. (2014), who described six groups (halter, western pleasure, reining, working cow, cutting, and racing short distances) within the breed. In Brazil, the first animals were imported from the state of Texas (USA) in 1955 and the Brazilian Association of Quarter Horse Breeders (ABQM) was founded in 1969. The Association registers the births of the animals and considers them to be of pure origin when their parents are registered in their Studbook or by the American Quarter Horse Association (AQHA). The Brazilian Studbook contains no information about the performance group of the animals, such as those cited by Petersen et al. (2014). Between 2012 and 2016, the ABQM recorded financial transactions in auctions throughout Brazil of about USD 302 million, with the sale of approxi- mately 27,000 animals. About 310,000 professionals directly work on horse farms distributed over an area of approximately 1 million hec- tares, with an estimated value of more than USD 6 billion (ABQM, 2017). Studies addressing important topics for the development and design of genetic breeding programs are necessary to guide technicians and breeders in the continuous improvement of the QH population. Parameters such as pedigree completeness (MacCluer et al., 1983), generation interval (GI), inbreeding coefficient (Wright, 1931) and the probability of gene origin (Boichard et al., 1997) are important to https://doi.org/10.1016/j.livsci.2018.06.001 Received 16 January 2018; Received in revised form 30 May 2018; Accepted 1 June 2018 ⁎ Corresponding author at: FMVZ - Unesp, DMNA - Fazenda Experimental Lageado, Rua José Barbosa de Barros, No. 1780, Botucatu, São Paulo, CEP: 18.618-307, Brasil. E-mail addresses: fariasky@gmail.com (R.A.S. Faria), rogcuri@fmvz.unesp.br (R.A. Curi), jaugusto@fmvz.unesp.br (J.A.I.V. Silva). Livestock Science 214 (2018) 135–141 1871-1413/ © 2018 Elsevier B.V. All rights reserved. T http://www.sciencedirect.com/science/journal/18711413 https://www.elsevier.com/locate/livsci https://doi.org/10.1016/j.livsci.2018.06.001 https://doi.org/10.1016/j.livsci.2018.06.001 mailto:fariasky@gmail.com mailto:rogcuri@fmvz.unesp.br mailto:jaugusto@fmvz.unesp.br https://doi.org/10.1016/j.livsci.2018.06.001 http://crossmark.crossref.org/dialog/?doi=10.1016/j.livsci.2018.06.001&domain=pdf design an animal breeding program. The data obtained permit to verify the genetic diversity and changes over time (Maignel et al., 1996). The objective of the study was to evaluate population parameters and to estimate the genetic diversity of QH in Brazil based on pedigree records. 2. Material and methods 2.1. Data and statistical analysis The pedigree data, including records of the animal, sire, dam, sex and date of birth, were provided by ABQM. Ancestors of sires not present in the file that are necessary to increase the quality of the pedigree were added. The data of these ancestors were obtained from the internet: Pedigree Online's All Breed Pedigree Database (www. allbreedpedigree.com). The total population (TP) consisted of 131,716 animals born between 1747 and 2008, with records of 60,933 males and 70,783 females born to 11,838 stallions and 34,028 mares. The reference population (RP) representing the last generation was applied in this study considering 47,861 animals born between 2000 and 2008, with records of 23,041 males and 24,820 females (36% of TP), born to 4503 stallions and 17,763 mares. Preparation of the data and statistical analysis were performed using the MEAN and FREQ procedures of the SAS software (SAS, 2011). The population and reproductive parameters, probability of gene origin and genetic diversity were obtained with the ENDOG V4.8 program (Gutiérrez and Goyache, 2005). 2.2. Reproductive parameters and generation interval The reproductive parameters including the mean, median, minimum, mode and maximum number of offspring, number of sires and mare-stallion ratio were calculated only for TP. The GI was ob- tained in both populations based on the average age of the parents at the birth of offspring that reproduced (James, 1972) and was calculated for the four different path of selection: sire-son, sire-daughter, dam-son, dam-daughter, and all parent-offspring. 2.3. Quality of pedigree data Pedigree completeness was calculated as the proportion of ancestors known in each ascending generation (MacCluer et al., 1983) and an- cestors with no known parent were considered founders as described by Gutiérrez and Goyache (2005). Providing equal information about the quality of pedigree data, the number of generations was computed in three different ways: (1) number of full generations traced (FG), which corresponds to the number of generations with both parents known; (2) number of equivalent complete generations (EG), which is calculated as the sum over all known ancestors based on (1/2)n, where n is the number of generations between the animal and each known ancestor, and best represents the pedigree information (Maignel et al., 1996); (3) maximum number of generations traced (MG), which provides the number of generations that separate the individual from its furthest ancestor, regardless of whether or not the parents are known. 2.4. Inbreeding parameters The individual inbreeding coefficient (F) is defined as the prob- ability that an individual has two identical alleles by descent (Wright, 1931) and was calculated using the algorithm proposed by Meuwissen and Luo (1992). The average relatedness coefficient (AR) of each an- imal was described by Goyache et al. (2003) and Gutiérrez et al. (2003) as the representation of a given animal in the pedigree and was defined as the probability that an allele randomly chosen from the population belongs to a given animal. The increase in inbreeding by generation (ΔF) was calculated with the classical formula = − −− −F F F FΔ ( )/(1 )t t t1 1 (1) where Ft and Ft-1 is the average inbreeding in the ith generation. To calculate the effective population size (Ne), defined by Gutiérrez and Goyache (2005) as the number of breeding animals that would lead to the actual increase in inbreeding if they contributed equally to the next generation, was considered: =N F1/2Δe (2) where ΔF indicates the average inbreeding increase per generation. In addition, Ne was calculated based on the individual increase in inbreeding as suggested by Gutiérrez et al. (2009) and was used for the calculation of genetic drift. 2.5. Probability of gene origin and genetic drift The effective number of founders (fe) is given by the measurement of the contributions of the most influential founders. Lacy (1989) de- fines fe as the number of equally contributing founders that would be expected to produce the same genetic diversity as in the population studied. This parameter was calculated using the formula ∑= = f q1/e k f k 1 2 (3) where qk is the probability of gene origin of ancestor k and f is the real number of founders. The effective number of ancestors (fa) is the minimum number of ancestors (founders or not) necessary to explain the genetic diversity of a population (Boichard et al., 1997). This parameter was calculated using the expression ∑= = f p1/a j f j 1 2 (4) where pj is the marginal contribution of ancestor j. The marginal con- tribution is the additional genetic contribution made by an ancestor that is not explained by another previously chosen ancestor (Boichard et al., 1997). The effective number of founder genomes (fg) is defined as the number of founders that would be expected to produce the same genetic diversity as in the population under study if the founders were equally represented and no loss of alleles occurred (Lacy, 1989). This parameter was estimated as proposed by Caballero and Toro (2000) using the formula =f C1/2g (5) where C is the average coancestry between individuals of the popula- tion. Genetic drift is the random change in allele frequencies in a popu- lation, which occurs at a higher intensity when the population under- goes a drastic reduction in its effective size (Falconer and Mackay, 1996). The ratio between fe/fa permits to identify whether the use of breeding animals causes a genetic bottleneck in the population (Boichard et al., 1997). A ratio equal to or close to 1 indicates the ab- sence of this effect. The ratio of fe/fg determines the effect of genetic drift in the population studied (Lacy, 1989). According to Sorensen et al. (1995), an fe/Ne ratio > 0.5 in a population indicates the oc- currence of changes in genetic drift. Stabilization of genetic drift in a population can be observed when fe is close to Ne/2, suggesting a greater representation of founders (Caballero and Toro, 2000). 2.6. Genetic conservation index The genetic conservation index (GCI) was proposed by Alderson (1992) and estimates the average effective number of founders in the pedigree of a given animal. Higher values indicate animals with a greater balance of genes transmitted by founders, i.e., greater genetic conservation of the breed. In addition, this index provides reliable R.A.S. Faria et al. Livestock Science 214 (2018) 135–141 136 http://www.allbreedpedigree.com http://www.allbreedpedigree.com predictions of homozygosity in subsequent generations. The GCI was calculated from the genetic contributions of all founders identified for each animal using the formula ∑= pGCI 1/ i 2 (6) where pi 2 is the proportion of genes of founder i in the pedigree of an animal (Gutiérrez and Goyache, 2005). 3. Results 3.1. Number of births over time The distribution of the official birth records (Fig. 1) includes only animals born after the opening (1969) of the Brazilian Studbook until 2008. In the first year, 104 births were recorded (18 males and 86 fe- males), with the number of animals increasing gradually thereafter. However, a smaller number of males than females were registered until 1980 (Fig. 1). After this year, the number of animals of both sexes in- creased constantly until 1993. The average annual growth rate was 19.7%, reaching 5785 records in 1993, followed by a decline between 1994 and 1999, and again increasing from 2000 to 2007 (5.2% per year). A marked decrease in the number of registered animals was observed in the last year studied (2008). 3.2. Reproductive parameters and generation interval Stallions and mares accounted for 19.4% and 48.1% of all animals, respectively. The average number of offspring per stallion and per mare in TP was 10.7 ± 29.9 and 3.7 ± 3.6, respectively (Table 1). Stallion Signed to fly and mare Princesa had the largest number of offspring (1006 and 186, respectively). The measures of central tendency (mode and median) of the number of offspring per breeding animal were the same and the ratio of the number of dams per sire in TP was 2.9. The average GI was 9.6 ± 5.3 and 10.8 ± 5.0 years for TP and RP, respectively (Table 2). The GI was greater for the sire-son than for the sire-daughter pathway and shorter for the dam-progeny than for the sire-progeny pathway. On average, mares reproduce at a younger age than stallions, indicating a greater contribution of females to the de- crease in this parameter. 3.3. Quality of pedigree data Pedigree completeness (Fig. 2) in the three most recent generations (1st, 2nd and 3rd), was 96.2%, 92.0% and 83.6% in TP and 100.0%, 100.0% and 98.5% in RP, respectively. The combined average of the five most recent generations was 77.9% and 91.0% for TP and RP, re- spectively. The results derived from the number of generations (Table 2) were 5.09 ± 1.76, 3.99 ± 1.0 and 26.76 ± 8.59 for EG, FG and MG, respectively, in TP. In the same order, these values were 6.24 ± 0.88, 4.34 ± 0.63 and 31.42 ± 4.63 in RP. In TP, the maximum number of FG and MG was 6 and 44 generations, respectively. 3.4. Inbreeding parameters The F was 1.07% and 1.56% in TP and RP, respectively, although 70.89% (TP) and 92.88% (RP) of the animals had an F different from zero (Table 3). The ΔF was also low (0.24%), with a small increase of Fig. 1. Number of animals born (males, female) and registered in the Brazilian Quarter Horse Studbook by year of birth. Table 1 Summary descriptive statistics of reproductive parameters of Quarter Horse stallions and mares in Brazil. Item Total population Stallion Mare Total animals (n) 60,933 70,783 Total breeding (n) 11,838 34,028 Total offspring (n) 126,311 126,083 Average ± SD (n) 10.7 ± 29.9 3.7 ± 3.6 Median (n) 2 2 Minimum and mode (n) 1 1 Maximum (n) 1006a 186b Mare/stallion ratio 2.9 n, number of the respective observations. SD, standard deviation a Signed to Fly. b Princesa. Table 2 Summary statistics of the generation interval (years) between parents and sons/ daughters that reproduced, and number of mean generations of the Quarter Horse breed in Brazil. Item Total population Reference population n Mean ± SD n Mean ± SD Generation interval Sire-son 10,090 10.0 ± 5.9 1384 11.5 ± 5.7 Sire-daughter 30,384 9.8 ± 5.5 3986 11.0 ± 4.9 Dam-son 10,055 9.3 ± 4.9 1384 10.5 ± 4.9 Dam-daughter 30,159 9.3 ± 4.8 3986 10.3 ± 4.6 Parents-offspring 80,688 9.6 ± 5.3 10,740 10.8 ± 5.0 Number of mean generation Equivalent complete (EG) – 5.09 ± 1.76 – 6.24 ± 0.88 Full (FG) – 3.99 ± 1.00 – 4.34 ± 0.63 Maximum (MG) – 26.76 ± 8.59 – 31.42 ± 4.63 n, number of observations. SD, standard deviation. Fig. 2. Completeness of pedigree information for both populations per gen- eration in Quarter Horses in Brazil. R.A.S. Faria et al. Livestock Science 214 (2018) 135–141 137 0.06% in RP. The Ne obtained based on ΔF (most common parameter in the literature) was 197 (TP) and 164 (RP). Using individual increase in inbreeding, the Ne was 199 and 167 for TP and RP, respectively (Table 3). 3.5. Probability of gene origin and genetic drift The three parameters of the probability of gene origin describe the genetic diversity of the populations (Table 3). The value of fe represents 19.3% and 16.8% of founders in TP and RP, respectively. Parameter fa had 889 ancestors less than fe in TP. The value of fg indicated that only 1.9% (TP) and 1.4% (RP) of the founder genomes are still represented in each population. Based on the total number of animals evaluated and when compared between each other, the parameters of the probability of gene origin, fe, fa and fg (Table 3), are considered low and indicate how many of the founders (1045) remain in the herd as ancestors (156) and founder genomes (105). The ratios of fe/fa and fe/fg were 6.70 and 9.95 (TP), and 7.18 and 12.29 (RP), respectively (Table 3). The fe/Ne ratio was 5.36 (TP) and 4.95 (RP), and the fe ∼ (Ne/2) value was 1045 ∼ 98 (TP) and 811 ∼ 82 (RP). 3.6. Genetic conservation index The mean GCI was 14.83 ± 8.89 and 21.55 ± 8.11 in TP and RP, respectively (Table 3). The indices lower than 2.0 and higher than 31.19 corresponded to the 5th and 95th percentiles, respectively (TP). There were 8023 animals with GCI values higher than 30. 3.7. Genetic contributions The total number of founders and ancestors of the QH breed in Brazil represent 4.1% and 3.7% (TP) and 10.1% and 6.3% (RP) of the animals, respectively, in each population (Table 3). The number of animals that explain 50% of the genetic diversity of the breed was small and represents only 2.5% and 2.4% of all ancestors in TP and RP, re- spectively. The cumulative genetic contribution of the 10, 100 and 500 most influential animals was increased in RP by 0.9%, 5.0% and 9.3% (founders) and by 5.3%, 8.7% and 7.6% (ancestors), respectively, compared to TP (Table 3). The marginal genetic contribution of the 15 most influential an- cestors of the QH breed was 21.52% and 27.34% in TP and RP, re- spectively (Table 4). The two most representative ancestors (Table 4) are Three Bars (Thoroughbred) and Doc Bar (Quarter Horse), together accounting for 7.78% (TP) and 9.36% (RP). These animals were born in the United States, had sired 51 offspring each, and breeding time for 22 and 16 years, respectively. The marginal genetic contribution of the most influential mare, Poco Lena, was 0.81% and 1.52% in TP and RP, respectively. The number of ancestors with a contribution greater than 1.0% was 8 and 13 animals in TP and RP, respectively. As also shown in Table 4, the mean breeding time was 14.1 ± 6.1 years in both populations, the mean number of offspring was 125 ± 129 and 170 ± 262 in TP and RP, respectively, and the mean EG was 5.65 ± 0.69 and 5.40 ± 0.73 in TP and RP, respectively. 4. Discussion 4.1. Number of births over time The decline in the number of births observed in the 1990s (Fig. 1) resembles that observed for autochthonous horse breeds in Brazil, in- cluding the Campolina (Procópio et al., 2003), Mangalarga Marchador (Costa et al., 2005), and Mangalarga breeds (Mota et al., 2006). This period coincided with the implementation of the Brazilian government's economic plan (currency change), suggesting a cause-effect relationship as reported by Mota et al. (2006). The decrease observed in 2008 (Fig. 1) coincided with the period of the financial crisis in Europe and in the United States that led to a deacceleration in horse breeding in dif- ferent parts of the world. A similar trend has been reported in studies on Lusitanos (Vicente et al., 2012) and Pantaneiros (McManus et al., 2013). Oscillations in the number of birth registrations observed in the 20th and 21st centuries are common in different continents and are influenced by economic-financial and social changes such as war. 4.2. Reproductive indices and generation interval The difference between stallion and mare in terms of the number of offspring was significant. Especially in the average and standard de- viation number of offspring per stallion (Table 1), indicates the con- centration of progeny in a reduced number of breeding animals, a finding suggesting the use of reproductive technologies. Differences in the mean number of offspring have been reported in the literature. Mota et al. (2006) reported a mean number of 23.8 per stallion and of 4.4 per dam for Mangalarga horses, Vicente et al. (2012) of 13.1 per stallion and of 4.1 per dam for Lusitano horses, and McManus et al. (2013) of 12.8 per stallion and of 2.6 per dam for the Pantaneiro breed. Table 3 Summary statistics of parameters related to inbreeding, probability of gene origin, genetic drift and genetic contributions of founders and ancestors of the Quarter Horse breed in Brazil. Item Total population Reference population Total number of animals in the study 131,716 47,861 Parameters related to inbreeding Average inbreeding coefficienta, F (%) 1.07 ± 2.72 1.56 ± 3.08 Average related coefficienta, AR (%) 0.96 ± 0.54 1.16 ± 0.41 Increase in inbreedinga, ΔF (%) 0.24 ± 0.82 0.30 ± 0.65 Animals with F different from zero (n) 93,359 44,456 Average inbreeding coefficientb, F (%) 1.51 ± 3.12 1.68 ± 3.16 Average related coefficientb, AR (%) 1.16 ± 0.44 1.20 ± 0.39 Increase in inbreedingb, ΔF (%) 0.33 ± 0.96 0.32 ± 0.67 Effective population sizec, Ne (n) 195 164 Effective population sized, Ne (n) 199 167 Probability of gene origin and genetic drift Effective number of founders, fe (n) 1045 811 Effective number of ancestors, fa (n) 156 113 Effective number of founder genomes, fg (n) 105 66 fe/fa 6.70 7.18 fe/fg 9.95 12.29 fe/Ne > 0.5 5.25 4.86 fe ∼ (Ne/2) 1045 ∼ 100 811 ∼ 84 Contributions of founders and ancestors Total number of animals with both parents known 125,913 47,861 Genetic conservation index (GCI) 14.83 ± 8.89 21.83 ± 8.11 Number of founders 5427 4815 Number of ancestors (100% of genetic diversity) 4821 2993 Number of ancestors (50% of genetic diversity) 121 72 Genetic contributions of the (%) 10 most influential founders 3.10 4.02 100 most influential founders 20.07 25.09 500 most influential founders 53.65 62.92 10 most influential ancestorse 17.45 22.48 100 most influential ancestorse 46.92 55.63 500 most influential ancestorse 74.86 82.48 a All animals in the population studied. b Only animals with an inbreeding coefficient ≠ from zero. c Gutiérrez and Goyache (2005). d Gutiérrez et al. (2009). e Marginal genetic contributions. R.A.S. Faria et al. Livestock Science 214 (2018) 135–141 138 The number of offspring per dam was low and varied little among the breeds studied, but contrasted with the mean number of offspring per stallion (Table 1). Reproductive differences between sexes are observed in all horse breeds, indicating a greater participation of dams in maintaining the genetic diversity of each breed. The largest number of offspring (Tables 1 and 4) was observed for stallion Signed to Fly and mare Princesa. The large number of offspring can be explained by the sportive results of breeding animals in terms of conformation traits, western modalities and/or speed racing. The GI was greater in RP (Table 2), reducing genetic gain probably because of the use of artificial insemination and/or embryo transfer, techniques that permit the reproductive use of deceased or older ani- mals. The high standard deviations indicate differences in the GI within the population. Long GI as obtained in the present study have been reported in the literature for different horse breeds. Petersen et al. (2014) found a similar mean GI (9.5 years) for QH in the United States. Vicente et al. (2012) reported a longer GI (10.5 years) for Lusitano horses. The GI is a parameter that is monitored in breeding programs because its reduction promotes an increase in genetic gain. However, the QH of most Brazilian breeders have long GI, reducing the rate of genetic improvements in their populations. 4.3. Quality of pedigree data Knowledge of the ancestors in the pedigree is necessary for char- acterizing the genetic diversity of breeds. According to (Cervantes et al., 2009), sets of incomplete data reduce the accuracy of the results. In the present study, in addition to the database of ABQM, the ancestors of Brazilian QH animals registered until the mid-18th century were used, permitting to evaluate the present genetic diversity of the QH breed in Brazil. Cervantes et al. (2008), Medeiros et al. (2014) and Duru (2017) reported pedigree completeness in the first, second and third genera- tions of 92.0%, 86.6% and 80.8% (Spanish Arab Horse in total popu- lation), 99.99%, 98.0% and 69.9% (Brazilian Sport Horse in reference population) and 98.2%, 96.6% and 95.0% (Turkish Arab Horse in re- ference population), respectively. Bartolomé et al. (2011), Roos et al. (2015) and Bhatnagar et al. (2011) described combined average pedi- gree completeness of the five most recent generations of 71.2% (Spanish Sport Horse in total population), 88.0% (Holstein in reference population) and approximately 70.0% (Norwegian Fjord Horse in North America), respectively. Similar to the literature, the increase in pedi- gree completeness in RP suggests improvement in the quality of the pedigree information of the ABQM registry (Fig. 2). The mean number of EG obtained (Table 2) permitted to evaluate the changes in genetic diversity that occurred in the last generation (RP). The values reported in the literature of 0.7 in Pantaneiro (McManus et al., 2013), of 5.7 in the Spanish Arab Horse (Cervantes et al., 2008) and of 9.9 in Lusitano horses (Vicente et al., 2012) indicate the difference in knowledge of pedigree information for each breed. The evolution of EG shows improvement in the pedigree information of each animal. However, in the presence of a breed founded in the 17th century, suggesting that the Brazilian Studbook of the QH breed should be expanded to include all ancestors from the United States, United Kingdom and other origins. Table 4 Summary statistics of the marginal genetic contribution (in %) of the 15 most influential ancestors in the total and reference population of the Quarter horse breed in Brazil, and year of birth, sex, breed, birth of first and last progeny, breeding time, total number of progeny and equivalent generations. Order Name Marginal contribution (%) Birth year Progeny Breeding time (years) EG (n) Birth of first Birth of last Total number Total population 1 Three Barsa 5.73 1940 1946 1968 51 22 5.75 2 Doc Bar 1.95 1956 1961 1977 51 16 5.09 3 Leo 1.76 1940 1946 1966 28 20 5.81 4 Top Decka 1.45 1945 1950 1966 21 16 4.60 5 El Zorrero 1.31 1969 1975 1993 382 18 5.48 6 Dash for Cash 1.11 1973 1979 1993 60 14 6.57 7 Mr San Peppy 1.10 1968 1974 1993 37 19 6.13 8 Sanjayc 1.05 1980 1982 1994 291 12 6.52 9 Shady Apolo Bars 0.99 1970 1973 1980 75 7 6.14 10 Dan's Boy Skippy 0.99 1969 1983 1993 311 10 6.02 11 Caracolito 0.91 1957 1961 1975 47 14 4.84 12 Catchme Ifyoucana 0.84 1975 1977 1997 306 20 5.66 13 Poco Lenab 0.81 1949 1967 1968 3 1 4.23 14 The Aquarian 0.75 1972 1977 1982 107 5 5.96 15 Easy Jet 0.74 1967 1970 1988 99 18 5.91 Reference population 1 Three Barsa 5.94 1940 1946 1968 51 22 5.75 2 Doc Bar 3.42 1956 1961 1977 51 16 5.09 3 Peppy San Badger 2.36 1974 1980 1993 94 13 5.29 4 Freckles Playboy 1.95 1973 1979 2000 55 21 5.43 5 El Zorrero 1.65 1969 1975 1993 382 18 5.48 6 Dash for Cash 1.60 1973 1979 1993 60 14 6.57 7 Poco Lenab 1.52 1949 1967 1968 3 1 4.23 8 Shady Apolo Bars 1.49 1970 1973 1980 75 7 6.14 9 Dan's Boy Skippy 1.45 1969 1983 1993 311 10 6.02 10 Leo San 1.10 1949 1956 1970 11 14 6.33 11 Easy Jet 1.07 1967 1970 1988 99 18 5.91 12 Peppy Belleb 1.06 1955 1961 1968 4 7 4.12 13 Trouble Two Times 1.04 1975 1978 2002 329 24 4.89 14 Beduinoa 0.88 1968 1976 1987 25 11 4.74 15 Signed to Fly 0.81 1988 1993 2008 1006 15 5.02 a Throughbred. b Female. c Born in Brazil. EG, equivalent generation. R.A.S. Faria et al. Livestock Science 214 (2018) 135–141 139 4.4. Inbreeding parameters Values of F (Table 3) have been reported in the literature, Bartolomé et al. (2011) and Hamann and Distl (2008) for the Spanish Sport Horse (0.66%) and Hanoverian horse (1.33%), respectively. High F values are observed mainly in the case of closed studbooks, for example Thor- oughbred (Cunningham et al., 2001), Lipizzaner (Zechner et al., 2002) and Lusitano (Vicente et al., 2012), contrary to the file used in this study. The low F observed in the two populations evaluated suggests that breeders control inbreeding in QH, but should be considered by breeders given the growing trend in the number of related animals. The low AR (Table 3) was similar to that described in the literature for other breeds (Bartolomé et al., 2011; Vicente et al., 2012; McManus et al., 2013), indicating reduced representation of each individual in the whole breed, regardless of pedigree completeness. Ne (ΔF) values of 372, 150, 226 and 55 have been reported for the Hanoverian (Hamann and Distl, 2008), Trakehner (Teegen et al., 2009), Spanish Sport Horse (Bartolomé et al., 2011) and Holstein (Roos et al., 2015) breeds, respectively. Studies describing high values of Ne suggest, by definition, reduced inbreeding because of the direct relationship between Ne and the inbreeding rate (Falconer and Mackay, 1996). However, the differences observed between the population studied and the literature are related to the number of animals in each breed and the probabilities of gene origin that maintain a close relationship with Ne. The FAO (1988) recommends a minimum Ne of 50. The value obtained in this study is not a matter of concern for the maintenance of genetic diversity, but the decrease in Ne observed in RP suggests the possible loss of genetic diversity. 4.5. Probability of gene origin and genetic drift The results of probability of gene origin reported in the literature are heterogeneous. Roos et al. (2015), Delgado et al. (2014) and Vicente et al. (2012) found lower values for Holstein horse (fe, 50; fa, 29 and fg, 17), purebred Arabian (fe, 30; fa, 13 and fg, 6), and Lusitano (fe, 28; fa, 12 and fg, 6), respectively. The higher values of the parameters obtained by Bartolomé et al. (2011) for the Spanish Sport Horse (fe, 963; fa, 407 and fg, 254) and by Siderits et al. (2013) for the German Paint Horse (fe, 963; fa, 186 and fg, 118) suggest the use of a larger number of animals for formation of the breeds. The diversity of the breeds observed in the literature, as well as of the probabilities of gene origin, suggests concern in the breeding of horses. However, the variation in the three para- meters of the probability of gene origin obtained in this study (Table 4) was higher than that described in the literature. Care should be taken when comparing the data with the literature because each population has its own characteristics. Comparison should be made within each breed and considering the three parameters of each population. The lower values of fe, fa and fg were found in RP compared to TP (Table 3), suggesting concentration of the population in a reduced number of breeding animals and loss of genetic diversity. The fe in both popula- tions, comparing the number of founders, indicates the preferential use of certain lineages of founders. Parameter fa complements the information provided by fe, in- dicating losses of genetic diversity produced by the unequal use of breeding animals, with the observation of high bottlenecks in both populations (Table 3). The results contrast with those obtained for the Holstein (1.72), purebred Arabian (2.31), Lusitano (2.33) and Spanish Sport Horse (2.37) breeds studied by Roos et al. (2015), Delgado et al. (2014), Vicente et al. (2012) and Bartolomé et al. (2011), respectively, who observed lower bottlenecks in the formation of the populations when compared to the present study. The smaller the difference be- tween fe and fa, the greater the number of founder animals in the po- pulation across generations, permitting the maintenance of the genetic base of the breed and of a large number of founders over time and consequently reducing the loss of genetic diversity, in contrast to what is observed in the QH breed. Values suggesting loss due to genetic drift have been reported for Hanoverians (Hamann and Distl, 2008) and the German Paint Horse (Siderits et al., 2013), but with smaller differences than those found in the present study (Table 3), indicating greater loss due to genetic drift in the QH breed in Brazil. 4.6. Genetic conservation index In the literature, mean GCI of 1.3 and 9.5 and maximum indices of 7.2 and 19.2 have been reported for Pantaneiro (McManus et al., 2013) and Lusitano horses (Vicente et al., 2012), respectively. The differences compared to the present study can be explained not only by the better conservation of the QH breed, but also by differences in the quality of the pedigree information. Animals with high values (GCI) exhibit greater genetic conservation, i.e., a greater balance in the number of founders, and should be used in genetic selection programs designed to maintain the genes transmitted by founders (Alderson, 1992). However, does not take into account pedigree bottlenecks, which may have oc- curred through non-founder individuals. These animals should be pre- ferred as breeding animals to ensure continuity of the genetic diversity of founders in the QH breed. 4.7. Genetic contributions Genetic contributions have been described by Hamann and Distl (2008), with a marginal contribution of the 15 top ancestors of up to 35% in the Hanoverian breed. Roos et al. (2015) found that 15 ances- tors (stallions) account for 50.0% of the total contribution, with a value of 11.6% for the most influential animals. Vicente et al. (2012) reported a marginal contribution of the main ancestor of 25% in the Lusitano breed. The differences found in the literature indicate that the QH ex- hibits greater genetic diversity in the 15 most influential ancestors. However, the increase in genetic contributions observed in RP (Tables 3 and 4) caused the loss of genetic diversity due to the preferential use of lineages of certain ancestors (founders or not). The long breeding time and large number of offspring of the main ancestors (Table 4) increase the concentration of genetic contributions in a small number of ancestors of the QH breed, indicating the loss of alleles in the last generation studied. The lower genetic contribution of females, indicating a greater genetic distribution compared to males, which therefore contribute to the maintenance of genetic diversity in QH. The values obtained do not compromise the evolution of the QH breed in Brazil. The breed continues to exhibit genetic diversity and a large number of ancestors in the formation of its population compared to the other breeds. However, a loss of genetic diversity could be ob- served (Tables 3 and 4) and may increase in future generations because of the lack of monitoring of the QH breed. Thus, the increase in genetic contributions must be monitored by ABQM and breeders to guide matings in order to obtain more balanced values and genetic diversity of the QH breed in Brazil. 5. Conclusion The results highlight the importance of monitoring the population structure in breeding programs, as well as of guided matings using ancestors of different lineages, in order to increase the genetic diversity of the QH breed in Brazil. The expectation of annual genetic gain in the population was reduced because of the long GI. Selection of a larger number of young sires with a shorter breeding time will permit to in- crease genetic gain, to reducing bottleneck effects and minimizing the loss of genetic diversity. Conflict of interest statement We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant R.A.S. Faria et al. Livestock Science 214 (2018) 135–141 140 financial support for this work that could have influenced its outcome. Acknowledgements The authors would like to thank the Associação Brasileira de Criadores do Cavalo Quarto de Milha (ABQM) for providing the data used in this study. We are also grateful to the anonymous reviewers for their helpful comments. Thanking the scholarship awarded to Ph.D. student Ricardo da Silva Faria, the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes - Brazil) and productivity grant at- tributed to Prof. J. Augusto II Silva, by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq - Brazil). References ABQM, A.B. dos criadores de cavalos Q. de M., 2017. Quarto de Milha no Brasil [WWW Document].. http://abqm.com.br/pt/conteudos/quarto-de-milha/quarto-de-milha- no-brasil/ (accessed 10.13.17). Alderson, L., 1992. A system to maximize the maintenance of genetic variability in small populations. In: Alderson, L., Bodo, I. (Eds.), Genetic Conservation of Domestic Livestock. CABI Publishing, Wallingford, pp. 18–29. Bartolomé, E., Cervantes, I., Valera, M., Gutiérrez, J.P., 2011. Influence of foreign breeds on the genetic structure of the Spanish Sport Horse population. Livest. Sci. 142, 70–79. http://dx.doi.org/10.1016/j.livsci.2011.06.021. Bhatnagar, A.S., East, C.M., Splan, R.K., 2011. Genetic variability of the Norwegian Fjord horse in North America. Anim. Genet. Resour. 49, 43–49. http://dx.doi.org/10.1017/ S2078633611000105. Boichard, D., Maignel, L., Verrier, É., 1997. The value of using probabilities of gene origin to measure genetic variability in a population. Genet. Sel. Evol. 29, 5–23. http://dx. doi.org/10.1186/1297-9686-29-1-5. Caballero, A., Toro, M., 2000. Interrelations between effective population size and other pedigree tools for the management of conserved populations. Genet. Res. 75, 331–343. http://dx.doi.org/10.1017/S0016672399004449. Cervantes, I., Gutiérrez, J.P., Molina, A., Goyache, F., Valera, M., 2009. Genealogical analyses in open populations: the case of three Arab-derived Spanish horse breeds. J. Anim. Breed. Genet. 126, 335–347. http://dx.doi.org/10.1111/j.1439-0388.2008. 00797.x. Cervantes, I., Molina, A., Goyache, F., Gutiérrez, J.P., Valera, M., 2008. Population his- tory and genetic variability in the Spanish Arab Horse assessed via pedigree analysis. Livest. Sci. 113, 24–33. http://dx.doi.org/10.1016/j.livsci.2007.02.011. Costa, M.D., Bergmann, J.A.G., Resende, A.S.C., Fonseca, C.G., 2005. Análise temporal da endogamia e do tamanho efetivo da população de eqüinos da raça Mangalarga Marchador. Arq. Bras. Med. Vet. e Zootec. 57, 112–119. http://dx.doi.org/10.1590/ S0102-09352005000100015. Cunningham, E.P., Dooley, J.J., Splan, R.K., Bradley, D.G., 2001. Microsatellite diversity, pedigree relatedness and the contributions of founder lineages to thoroughbred horses. Anim. Genet. 32, 360–364. http://dx.doi.org/10.1046/j.1365-2052.2001. 00785.x. Delgado, J.F., De Andrés, N., Valera, M., Gutiérrez, J.P., Cervantes, I., 2014. Assessment of population structure depending on breeding objectives in Spanish Arabian horse by genealogical and molecular information. Livest. Sci. 168, 9–16. http://dx.doi.org/ 10.1016/j.livsci.2014.07.012. Duru, S., 2017. Pedigree analysis of the Turkish Arab horse population: structure, in- breeding and genetic variability. Animal 11, 1449–1456. http://dx.doi.org/10.1017/ S175173111700009X. Falconer, D., Mackay, T., 1996. Introduction to Quantitative Genetics, fourth ed. Longmans Green, Harlow, Essex, UK, pp. 464. FAO, 1988. Secondary Guidelines for Development of National Farm Animal Genetic Resources Management Plans: Management of Small Populations at Risk. Food and Agricultural Organization, Roma, Italy. Goyache, F., Gutiérrez, J.P., Fernández, I., Gomez, E., Alvarez, I., Díez, J., Royo, L.J., 2003. Using pedigree information to monitor genetic variability of endangered po- pulations: the Xalda sheep breed of Asturias as an example. J. Anim. Breed. Genet. 120, 95–105. http://dx.doi.org/10.1046/j.1439-0388.2003.00378.x. Gutiérrez, J.P., Altarriba, J., Díaz, C., Quintanilla, R., Cañón, J., Piedrafita, J., 2003. Pedigree analysis of eight Spanish beef cattle breeds. Genet. Sel. Evol. 35, 43–63. http://dx.doi.org/10.1051/gse. Gutiérrez, J.P., Cervantes, I., Goyache, F., 2009. Improving the estimation of realized effective population sizes in farm animals. J. Anim. Breed. Genet. 126, 327–332. http://dx.doi.org/10.1111/j.1439-0388.2009.00810.x. Gutiérrez, J.P., Goyache, F., 2005. A note on ENDOG: A computer program for analysing pedigree information. J. Anim. Breed. Genet. 122, 172–176. http://dx.doi.org/10. 1111/j.1439-0388.2005.00512.x. Hamann, H., Distl, O., 2008. Genetic variability in Hanoverian warmblood horses using pedigree analysis. J. Anim. Sci. 86, 1503–1513. http://dx.doi.org/10.2527/jas.2007- 0382. James, J.W., 1972. Computation of genetic contributions from pedigrees. Theor. Appl. Genet. 42, 272–273. http://dx.doi.org/10.1007/BF00277555. Lacy, R.C., 1989. Analysis of founder representation in pedigrees - founder equivalents and founder genome equivalents. Zoo Biol. 8, 111–123. http://dx.doi.org/10.1002/ zoo.1430080203. MacCluer, J.W., Boyce, A.J., Dyke, B., Weitkamp, L.R., Pfennig, D.W., Parsons, C.J., 1983. Inbreeding and pedigree structure in Standardbred horses. J. Hered. 74, 394–399. Maignel, L., Boichard, D., Verrier, E., 1996. Genetic variability of French dairy breeds estimated from pedigree information. Interbull 14, 49–54. McManus, C., Santos, S.A., Lima Dallago, B.S., Paiva, S.R., Saraiva Martins, R.F., Neto, J.B., Marques, P.R., Pinto de Abreu, U.G., 2013. Evaluation of conservation program for the pantaneiro horse in Brazil. Rev. Bras. Zootec. 42, 404–413. http://dx.doi.org/ 10.1590/S1516-35982013000600004. Medeiros, B.R., Bertoli, C.D., Garbade, P., McManus, C.M., 2014. Brazilian Sport Horse: pedigree analysis of the Brasileiro de Hipismo breed. Ital. J. Anim. Sci. 13, 657–664. http://dx.doi.org/10.4081/ijas.2014.3146. Meuwissen, the, Luo, Z., 1992. Computing inbreeding coefficients in large populations. Genet. Sel. Evol. 24, 305–313. http://dx.doi.org/10.1186/1297-9686-24-4-305. Mota, M.D.S., Prado, R.S.A., Sobreiro, J., 2006. Caracterização da população de cavalos Mangalarga no brasil. Arch. Zootec. 55, 31–37. Petersen, J.L., Mickelson, J.R., Cleary, K.D., McCue, M.E., 2014. The american quarter horse: Population structure and relationship to the thoroughbred. J. Hered. 105, 148–162. http://dx.doi.org/10.1093/jhered/est079. Procópio, A.M., Bergmann, J.A.G., Costa, M.D., 2003. Formação e demografia da raça Campolina. Arq. Bras. Med. Vet. e Zootec. 55, 361–365. http://dx.doi.org/10.1590/ S0102-09352003000300018. Roos, L., Hinrichs, D., Nissen, T., Krieter, J., 2015. Investigations into genetic variability in Holstein horse breed using pedigree data. Livest. Sci. 177, 25–32. http://dx.doi. org/10.1016/j.livsci.2015.04.013. SAS, 2011. SAS/STAT 9.3. User's Guid. SAS Inst. Inc., Cary, NC. Siderits, M., Baumung, R., Fuerst-Waltl, B., 2013. Pedigree analysis in the German Paint Horse: Genetic variability and the influence of pedigree quality. Livest. Sci. 151, 152–157. http://dx.doi.org/10.1016/j.livsci.2012.10.018. Sorensen, D., Andersen, S., Gianola, D., Korsgaard, I., 1995. Bayesian inference in threshold models using Gibbs sampling. Genet. Sel. Evol. 27, 229–249. http://dx.doi. org/10.1186/1297-9686-27-3-229. Teegen, R., Edel, C., Thaller, G., 2009. Population structure of the Trakehner Horse breed. Animal 3, 6–15. http://dx.doi.org/10.1017/S1751731108003273. Vicente, A.A., Carolino, N., Gama, L.T., 2012. Genetic diversity in the Lusitano horse breed assessed by pedigree analysis. Livest. Sci. 148, 16–25. http://dx.doi.org/10. 1016/j.livsci.2012.05.002. Wright, S., 1931. Evolution in Mendelian populations. Genetics 16, 97–159. http://dx. doi.org/10.1016/S0092-8240(05)80011-4. Zechner, P., Sölkner, J., Bodó, I., Druml, T., Baumung, R., Achmann, R., Marti, E., Habe, F., Brem, G., 2002. Analysis of diversity and population structure in the Lipizzan horse breed based on pedigree information. Livest. Prod. Sci. 77, 137–146. http://dx. doi.org/10.1016/S0301-6226(02)00079-9. R.A.S. Faria et al. Livestock Science 214 (2018) 135–141 141 http://abqm.com.br/pt/conteudos/quarto-de-milha/quarto-de-milha-no-brasil/ http://abqm.com.br/pt/conteudos/quarto-de-milha/quarto-de-milha-no-brasil/ http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0002 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0002 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0002 http://dx.doi.org/10.1016/j.livsci.2011.06.021 http://dx.doi.org/10.1017/S2078633611000105 http://dx.doi.org/10.1017/S2078633611000105 http://dx.doi.org/10.1186/1297-9686-29-1-5 http://dx.doi.org/10.1186/1297-9686-29-1-5 http://dx.doi.org/10.1017/S0016672399004449 http://dx.doi.org/10.1111/j.1439-0388.2008.00797.x http://dx.doi.org/10.1111/j.1439-0388.2008.00797.x http://dx.doi.org/10.1016/j.livsci.2007.02.011 http://dx.doi.org/10.1590/S0102-09352005000100015 http://dx.doi.org/10.1590/S0102-09352005000100015 http://dx.doi.org/10.1046/j.1365-2052.2001.00785.x http://dx.doi.org/10.1046/j.1365-2052.2001.00785.x http://dx.doi.org//10.1016/j.livsci.2014.07.012 http://dx.doi.org//10.1016/j.livsci.2014.07.012 http://dx.doi.org/10.1017/S175173111700009X http://dx.doi.org/10.1017/S175173111700009X http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0013 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0013 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0014 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0014 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0014 http://dx.doi.org/10.1046/j.1439-0388.2003.00378.x http://dx.doi.org/10.1051/gse http://dx.doi.org/10.1111/j.1439-0388.2009.00810.x http://dx.doi.org/10.1111/j.1439-0388.2005.00512.x http://dx.doi.org/10.1111/j.1439-0388.2005.00512.x http://dx.doi.org/10.2527/jas.2007-0382 http://dx.doi.org/10.2527/jas.2007-0382 http://dx.doi.org/10.1007/BF00277555 http://dx.doi.org/10.1002/zoo.1430080203 http://dx.doi.org/10.1002/zoo.1430080203 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0023 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0023 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0024 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0024 http://dx.doi.org/10.1590/S1516-35982013000600004 http://dx.doi.org/10.1590/S1516-35982013000600004 http://dx.doi.org/10.4081/ijas.2014.3146 http://dx.doi.org/10.1186/1297-9686-24-4-305 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0028 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0028 http://dx.doi.org/10.1093/jhered/est079 http://dx.doi.org/10.1590/S0102-09352003000300018 http://dx.doi.org/10.1590/S0102-09352003000300018 http://dx.doi.org/10.1016/j.livsci.2015.04.013 http://dx.doi.org/10.1016/j.livsci.2015.04.013 http://refhub.elsevier.com/S1871-1413(18)30178-1/sbref0032 http://dx.doi.org/10.1016/j.livsci.2012.10.018 http://dx.doi.org/10.1186/1297-9686-27-3-229 http://dx.doi.org/10.1186/1297-9686-27-3-229 http://dx.doi.org/10.1017/S1751731108003273 http://dx.doi.org/10.1016/j.livsci.2012.05.002 http://dx.doi.org/10.1016/j.livsci.2012.05.002 http://dx.doi.org/10.1016/S0092-8240(05)80011-4 http://dx.doi.org/10.1016/S0092-8240(05)80011-4 http://dx.doi.org/10.1016/S0301-6226(02)00079-9 http://dx.doi.org/10.1016/S0301-6226(02)00079-9 Assessment of pedigree information in the Quarter Horse: Population, breeding and genetic diversity Introduction Material and methods Data and statistical analysis Reproductive parameters and generation interval Quality of pedigree data Inbreeding parameters Probability of gene origin and genetic drift Genetic conservation index Results Number of births over time Reproductive parameters and generation interval Quality of pedigree data Inbreeding parameters Probability of gene origin and genetic drift Genetic conservation index Genetic contributions Discussion Number of births over time Reproductive indices and generation interval Quality of pedigree data Inbreeding parameters Probability of gene origin and genetic drift Genetic conservation index Genetic contributions Conclusion Conflict of interest statement Acknowledgements References