Genetic relationship between free movement and under rider gaits in young Pura Raza Española horses

Livestock Science 263 (2022) 105031 Available online 15 July 2022 1871-1413/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Genetic relationship between free movement and under rider gaits in young Pura Raza Espa˜ nola horses M. Ripoll´ es-Lobo * , D.I. Perdomo-Gonz´ alez, M.J. S´ anchez-Guerrero, E. Bartolom´ e, M. Valera Departamento de Agronomía, ETSIA, Universidad de Sevilla, Ctra. Utrera Km 1, Sevilla 41013, Spain HIGHLIGHTS •An early preselection in young horses saves costs and increases the genetic progress in the improvement of the range of functions of the PRE horse. •The heritability for the three natural gaits, can be considered suitable for genetic selection of breeding stock based on genetic values. •The high genetic correlations found between the same gaits according to the type of presentation of the young horse allow us to carry out an early pre-selection that can participate in dressage competitions. ARTICLE INFO Keywords: Basic aptitude Dressage equine Gaits traits Heritability ABSTRACT The Pura Raza Espa˜ nola (PRE) horse is an autochthonous Spanish horse characterized by its versatility, mental balance, and its balanced conformation, which makes it suitable for sports disciplines such as dressage. The aim of this study was to determine the genetic correlation between the quality of natural gaits in young horses, when the horse is evaluated in the Basic Aptitude Test (free movements, FM) and during dressage competitions (functionality tests and gaits under rider, UR). A total of 37,822 and 2730 PRE horses were evaluated for FM and UR, respectively, of which 1023 horses had records for both (coincidence animals). Genetic parameters were estimated using VCE software. Effects were included in the final model according to significant results in analyses of variance (GLM). For FM and UR, the genetic model included age as a linear covariate, gender, coat colour and stud size as fixed effects, and for UR, the rider and permanent environmental effects were also included as random effects. The pedigree file information included a minimum of 4 generations: 88,558 horses (FM), 13,837 horses (UR) and 9072 horses (coincidence animals). Heritabilities were estimated for the two data sets (FM and UR) and, for the coincidence population, heritabilities and genetic correlations were also estimated. The heritability for walk, trot and canter (FM) were 0.20, 0.22, 0.22, while for UR they were 0.26, 0.31, 0.33, respectively. For coincidence animals, the heritability for walk, trot and canter (FM) were 0.43, 0.33, 0.22, while for UR they were 0.23, 0.11, 0.10, respectively. The genetic correlations between FM and UR traits were high and positive; 0.49 for walk, 0.61 for trot and 0.54 for canter. The genetic parameters obtained highlight the possibility of preselecting gait traits in young horses destined to participate in sporting competitions. 1. Introduction The horse is an athletic animal whose economic value depends mainly on its sports performance (Wallin et al., 2003; Braam et al., 2011). However, such performance is the result of a complex combination of training, and conformational, physiological and behavioural characteristics which are largely inherited (Giulotto, 2001; Brooks et al., 2010; Sol´ e et al., 2013; Sanchez et al., 2016a). In sport horse breeding programs, great importance has always been attached to improving gait quality, despite the differences between certain breeds (Barrey et al., 2002; Koenen et al., 2004; Komosa et al., 2013; S´ anchez et al., 2014). Most countries select horses using conformation as an indication of later performance. Horses with a favourable conformation are assumed to perform better (Wallin et al., 2003). For this reason, there is a growing need for studying and accurately examining functionality and kinematic characteristics in most horse breeds, as in the case of the following breeds: Swedish Warmblood horses (Wallin et al., 2003), Warmblood sport Horses (Koenen et al., 2004), Lipizzaner (Baban et al., 2009), * Corresponding author. E-mail address: [email protected] (M. Ripoll´ es-Lobo). Contents lists available at ScienceDirect Livestock Science journal homepage: www.elsevier.com/locate/livsci https://doi.org/10.1016/j.livsci.2022.105031 Received 24 May 2022; Received in revised form 6 July 2022; Accepted 13 July 2022 Livestock Science 263 (2022) 105031 2 Jumping horses (Lewczuk and Ducro, 2012), Menorca horses (Sol´ e et al., 2013), Lusitano (Sol´ e et al., 2014), Campolina horses (Bussiman et al., 2018), French trotters (G´ orniak et al., 2020), Icelandic horses (Rosengren et al., 2021), and the Pura Raza Espa˜ nola (PRE) horse (Molina et al., 2008; Valera et al., 2008; Sanchez et al., 2013a; 2016a). The PRE is the most important equine breed in Spain from the census point of view and one of those that generates the greatest impact on economy and tourism throughout the country (S´ anchez-Guerrero et al., 2017a). Its characteristics make it a very versatile and functional animal, and it is one of the most valued equine breeds worldwide (Valera, 2017; Poyato-Bonilla et al., 2018), present in over 65 countries (Sol´ e et al., 2018). The main objective of the PRE breeding program is to obtain animals with favourable functional aptitudes that enable outstanding performance in dressage competitions, for this reason, gait quality has a considerable impact on the value of a horse. Dressage is a discipline based on the harmony between rider and horse, and requires a lot of training to perform difficult movements. Here, the horse’s gaits are of major relevance, since they determine the final score obtained by the animals participating in these events. For this reason, the rider has been considered a relevant environmental factor in the genetic evaluations of dressage horses (Kearsley et al., 2008; Viklund et al., 2011; S´ anchez et al., 2014, 2016b). The study of the natural gaits of PRE horses in free movement is performed by assessing the basic aptitudes of young horses at the age of at least three years old. This basic aptitude assessment is an essential requirement for the horses to be registered in the breed studbook. The main objective of this study was to determine the genetic correlation between the quality of the natural gaits in young horses, when the horse’s basic aptitude is evaluated (free movements, FM) during functionality tests and gaits under rider (UR), during dressage competitions. 2. Material and methods 2.1. Data set The functional variables analysed correspond to the three gaits of the horse: walk, trot, and canter. In basic aptitude, each gait was evaluated separately by 1 judge, who gave the horse a score ranging from 1 to 9 points, on a linear scale where 1 was the worst mark and 9 the best (S´ anchez et al., 2013a). These evaluations were carried out by a total of ten judges, all of whom had been previously trained and tested, and had the same evaluation criteria. In dressage competitions, the score used was the average from the three judges for each gait trait (the total number of judges for dressage was 77). All the judges involved were international judges recognized by the Royal Spanish Equestrian Federation (RFHE). In addition, all the competitions were connected through the judge effect (repetition of judges in the different competitions). The FM data consisted of 37,822 records of walk, trot, and canter from 37,822 PRE horses (12,752 males and 25,070 females) with an average age of 4.73 years old. The data was collected from 2012 to 2021 by the Royal Association of Spanish Horse Breeders (ANCCE), during basic aptitude assessment of the PRE for registration in the stud book of potential breeding stock. The dressage data consisted of 13,694 records of walk, trot, and canter from 2730 PRE horses (2559 males and 171 females), with an average age of 4.92 years old. The data was collected from 2004 to 2021 by the ANCCE and the RFHE during dressage reprises. To participate in the performance test (dressage and basic aptitude), the horse had to be registered in the Pura Raza Espa˜ nola horse studbook. A selection of coincidence horses was made according to the following criteria: - Horses participating in FM and UR. - Horses ridden by more than one rider. - Horses that had participated at least three times in dressage. - Age between 4 and 6 years old. The number of coincidence horses that met these criteria was 1023 (931 males and 92 females), with a total of 8142 records for UR (Table 1). 2.2. Genetic model A General Linear Model was used to examine the associations between gaits (walk, trot and canter) and the potential effects. A Tukey Post-hoc analysis was also carried out to analyse the significant differences between factor levels. Statistical analyses were performed using SAS (Statistical Analysis System, version 9.2, 2011). The genetic parameters were estimated with VCE software (Groeneveld, 2010) which allows us to analyse two different datasets with two different models at the same time, one for FM and other for UR. The analysis was performed with three different approaches: (1) Univariate genetic analysis for FM; (2) Univariate genetic analysis for UR; and (3) Bivariate genetic analysis with heterogeneity of variance for coincidence animals (FM+UR). The equations in matrix notation to solve the mixed model were: YFM =Xb +Zu +e(1) The effects studied for FM were: age as linear covariate (expressed in months), gender (2 levels; male and female), coat colour (4 levels; bay, chestnut, black, and grey), and stud size (5 levels;1–2 foals born per year, 3–6 foals born per year, 7–12 foals born per year, >13 foals born per year, and non-breeding livestock). YUR =Xb +Zu +Qr +Wp +e(2) The effects studied for UR were: age as linear covariate (expressed in months), gender (2 levels; male and female), coat colour (4 levels; bay, chestnut, black, and grey), and stud size (5 levels;1–2 foals born per year, 3–6 foals born per year, 7–12 foals born per year, >13 foals born per year, and non-breeding livestock). The two random effects also studied were rider (1113 levels) and permanent environmental effect (2730 levels). YFM =Xb +Zu +e(3) YUR =Xb +Zu +Qr +Wp +e The effects studied for the coincidence animals were: age as linear covariate (expressed in months), gender (2 levels; male and female), coat colour (4 levels; bay, chestnut, black, and grey), and stud size (5 levels;1–2 foals born per year, 3–6 foals born per year, 7–12 foals born per year, >13 foals born per year, and non-breeding livestock). The two additional random effects studied were rider (565 levels) and permanent environmental effect (1023 levels). where u∼N(0,A σ 2 u), r∼N(0,I σ 2 r), p∼N(0,I σ 2 p)and e∼N(0,I σ 2 e); y =vector of observations (phenotypic values); b =vector of systematic effects; u=vector of infinitesimal additive genetic contribution; r =vector of rider effects; p =vector of permanent environmental effect; e=vector of residuals; X=incidence matrix of systematic effects; Z=incidence matrix of additive genetic contribution; Q =incidence matrix of the rider effect and W =incidence matrix of permanent environmental effects. σ 2 u =direct genetic variance; σ 2 r =rider variance; σ 2 p =permanent environmental variance; σ 2 e = Table 1 Number of participants and records of Pura Raza Espa˜ nola horses during dressage competitions and basic aptitude test. Animals Male Female Records Basic aptitude 37,822 12,752 25,070 37,822 Dressage test 2,730 2,559 171 13,694 Coincidence horses 1,023 931 92 8,142 M. Ripoll´ es-Lobo et al. Livestock Science 263 (2022) 105031 3 residual variance ; A matrix of additive genetic correlations between individuals, I the identity matrix that links all individuals in the pedigree. The pedigree information for the genetic evaluation was collected from the PRE official studbook. At least four generations of all the horses in the control group were included in the pedigree file. For each of the assessments (FM, UR, and coincidence animals), a pedigree file was built with 88,558, 13,837 and 9072 individuals, respectively. 3. Results The descriptive statistics of the FM and UR gaits analysed in PRE horses are shown in Table 2. The average values oscillated between 6.37 (canter-FM) and 6.74 (canter-UR), with a coefficient of variation of 17.62% and 7.68%, respectively. The highest mean value for FM was obtained for trot (6.56) while in UR the highest value was obtained for canter (6.74). It was also observed that the average values in FM were always lower than those obtained in UR, while the lowest coefficient of variation was obtained for UR. The minimum value obtained in the evaluation of both disciplines was 1 and the maximum 9; the trot was the gait which was awarded the most points in the horses under rider in dressage competitions (Table 2). The general linear model results are shown in Table 3. All the factors studied in FM (gender, age, coat colour, stud size) and in UR (gender, coat colour, stud size, age, rider and permanent environmental effect) showed statistically significant differences for all the traits analysed. The genetic parameters for FM and UR movements are shown in Table 4. The heritability ratios showed moderate values for all the variables, oscillating from 0.20 (walk) to 0.22 (canter) for FM and from 0.26 (walk) to 0.33 (canter) for UR. For coincident horses, heritability for FM oscillated between 0.22 (canter) to 0.43 (walk) and for UR from 0.10 (canter) to 0.23 (walk). The genetic correlations between FM and UR, estimated from coincident animals, were of a high magnitude, ranging from 0.49 for walk to 0.61 for trot. Fig. 1 shows the percentage of coincident animals within 25% and 10% of the highest estimated breeding values (EBV) for each of the three gaits (walk, trot and canter). As can be seen, within 25% of the highest EBV, the percentage of coincident animals oscillates between 76.76% (trot) and 62.43% (walk). Meanwhile, the coincident animals within the 10% highest EBV oscillates between 68.36% (trot) and 52.04% (walk). 4. Discussion In sport horses, gait quality has a considerable influence on their functional performance, especially in a sport discipline such as dressage. Animals that are balanced and elastic are highly sought after, and these qualities score highly at all levels of competition (Becker et al., 2011). However, the PRE horse has its own functional aptitudes and, therefore, certain characteristics of its three natural gaits, which differentiate it from most equine sport breeds. In selection programs, it is vital to evaluate the animals at an early stage, which allows animals to be selected for training for later participation in sporting activities (G´ omez-Ortiz et al., 2021). Evaluations of young horses in sport competitions have been used successfully in many breeds recognized by the World Breeding Federation of Sport Horses (WBFSH), such as Swedish (Viklund et al., 2008; J¨ onsson et al., 2014; Viklund and Eriksson, 2018), Dutch (Ducro et al., 2007a, 2009), Belgian (Rustin et al., 2009) Warmbloods and Hanoverian (Schr¨ oeder et al., 2010), Polish (Borowska et al., 2011), British sport horses (Stewart et al., 2011) and German horses (Sch¨ opke et al., 2013). Good scores in gait quality are essential both in the basic aptitude evaluation tests (FM) and in dressage competitions (UR), and for this Table 2 Descriptive statistics analysis of natural gait traits evaluated in free movement during evaluation of basic aptitude in Pura Raza Espa˜ nola horses and under the rider during evaluation of dressage competitions. Comparison of means (Student’s t-test) between horse gaits. Trait Free movement Under Rider p Mean ±SE C.V. (%) Min. Max. Mean ±SE C.V. (%) Min. Max. Walk 6.46 ±0.004 14.89 1 9 6.60 ±0.005 9.26 2 9 *** Trot 6.56 ±0.005 16.49 1 9 6.68 ±0.005 8.19 1 9 *** Canter 6.37 ±0.006 17.62 1 9 6.74 ±0.004 7.68 1 9 *** C.V. (%) =Coefficient of variation; SE. standard error of the mean; Max=maximum; Min=minimum; *** p<0.001. Table 3 Means ±standard errors of mean, General linear model (GLM) between the gaits (walk, trot and canter) with the different factors analysed and Post-Hoc (Tukey). Factors Free movement Under rider Walk Trot Canter Walk Trot Canter Gender *** *** *** * *** n.s. Male 6.56 ± 0.009 b 6.65 ± 0.009 b 6.65 ± 0.010 b 6.40 ± 0.005 a 6.51 ± 0.005 a 6.61 ± 0.004 Female 6.42 ± 0.005 a 6.51 ± 0.006 a 6.22 ± 0.007 a 6.45 ± 0.020 b 6.59 ± 0.018 b 6.61 ± 0.016 Coat color *** *** *** n.s. *** *** Bay 6.45 ± 0.008 b 6.52 ± 0.009 b 6.35 ± 0.009 b 6.43 ± 0.008 6.58 ± 0.007 b 6.61 ± 0.007 bc Chestnut 6.32 ± 0.019 a 6.42 ± 0.022 a 6.29 ± 0.023 ab 6.47 ± 0.029 6.55 ± 0.027 ab 6.67 ± 0.025 b Black 6.38 ± 0.014 a 6.45 ± 0.015 a 6.28 ± 0.016 a 6.38 ± 0.018 6.52 ± 0.015 a 6.59 ± 0.015 ac Grey 6.53 ± 0.007 c 6.63 ± 0.008 c 6.41 ± 0.008 c 6.43 ± 0.007 6.54 ± 0.006 a 6.56 ± 0.006 a Stud size *** *** *** ** ** ** 1–2 foals per year 6.42 ± 0.008 a 6.46 ± 0.009 a 6.26 ± 0.010 a 6.43 ± 0.011 ab 6.57 ± 0.010 b 6.63 ± 0.010 b 3–6 foals per year 6.50 ± 0.010 b 6.60 ± 0.012 b 6.40 ± 0.012 b 6.46 ± 0.011 b 6.56 ± 0.010 b 6.62 ± 0.009 ab 7–12 foals per year 6.49 ± 0.013 b 6.65 ± 0.015 b 6.43 ± 0.016 b 6.42 ± 0.013 ab 6.55 ± 0.013 ab 6.62 ± 0.012 ab >13 foals per year 6.55 ± 0.013 c 6.73 ± 0.014 c 6.55 ± 0.014 c 6.40 ± 0.012 a 6.54 ± 0.010 ab 6.59 ± 0.009 ab Non-breeding livestock 6.40 ± 0.013 a 6.44 ± 0.014 a 6.31 ± 0.015 a 6.43 ± 0.010 ab 6.52 ± 0.008 a 6.58 ± 0.009 a Age *** *** *** *** *** *** Rider – – – *** *** *** Permanent environmental effect – – – *** *** *** Different superscript letters (a, b, c, d, e) indicate statistically significant difference between means at ***p<0.001; ** p<0.01; * p<0.05. M. Ripoll´ es-Lobo et al. Livestock Science 263 (2022) 105031 4 reason, these variables are included in the performance controls of the PRE breeding program. In our study, we found gait scores in PRE horses (FM and UR) of between 6 and 7 points, as well as in Swedish Warmblood horses (Viklund et al., 2008, 2010) and Hispano-Arabian horses (G´ omez et al., 2016). Barrey et al. (2002) compared the gaits of walk and trot in German horses (Hanoverian, Oldenburger, and Westfalian), French saddle horses (Selle Français), and PRE horses, to evaluate their aptitude for dressage, with the German horses obtaining higher scores for their gaits. In our study, the gaits traits of the Pura Raza Espa˜ nola horse usually obtained higher scores when the horse was presented UR than for FM. This is probably due to the fact that the horses which participate in dressage have better quality gaits, thanks to their continuous training and because the best animals have been pre-selected to participate in this type of sporting event. Regardless of the breed, most sport horse breeders aim to breed horses that excel in sport competitions, mainly dressage or show jumping, and different methods have been adopted across Europe to evaluate the functional performances of these horses (Bruns, 2001). To achieve good genetic progress of the horse in the dressage discipline, the intensity of selection must be high, as well as the heritability of the traits evaluated in competition (Hellsten et al., 2006). In this regard, a reliable estimation of genetic parameters (heritability and correlations) is essential to plan and carry out any selection programme. There are many factors that influence horse gaits (S´ anchez et al., 2014). The study of the environmental factors (Table 3) has shown that males are more successful in basic aptitude scores, while females score higher in dressage. In contrast, with Swedish Warmblood horses, males score higher in dressage tests (Viklund et al., 2008, 2010). However, most of the articles that study horse gaits do not differentiate the score by sex (Albertsd´ ottir et al., 2008; Becker et al., 2011; S´ anchez-Guerrero et al., 2017a). Another effect that has been found to be significant is the coat, coinciding with S´ anchez-Guerrero et al. (2019), who found significant differences in numerous conformation variables as a function of coat colour in the PRE horse or with the study by Perdomo-Gonz´ alez et al. (2022) in the Pura Raza Menorquina horse. An association has also been found in the PRE between coat colour and certain conformational defects such as cresty neck (S´ anchez et al., 2017b) or deep neck (Ripoll´ es et al., 2020). The genetic correlations between conformation and gaits are usually reported to be positive and moderate to high (Ducro et al., 2007b; Vicente et al., 2014; Sanchez-Guerrero et al., 2017a). In this study, horses with a grey coat obtained higher scores for gaits in FM, compared to horses with other coats, although chestnut and bay horses showed the best scores in UR. According to S´ anchez-Guerrero et al. (2019), PRE bay horses showed the greatest height at withers. This variable is perhaps one of the most studied, because of its relationship with the functionality of the horse, showing a positive relationship with the horse’s sport performance (Magnusson and Thafvelin, 1985; S´ anchez-Guerrero et al., 2017a, 2019) or the scores obtained in morphological contests (Holmstr¨ om and Philipsson, 1993). Galisteo et al. (1998) estimated significant and high correlations with stride length and length of overtracking and a moderate influence with angular parameters. In this context, S´ anchez et al. (2013b) showed that withers height had a significant relationship with most of the biokinematic variables (14 variables: 6 temporal, 4 linear and 4 angular), with a 0.55 relationship with hindlimb length. Stud size is extremely important, because it influences the size of the farm and the type of management carried out. As observed in this study, horses from large studs, preferably dedicated to breeding and reproduction, showed higher FM gait scores, probably because these horses are under extensive management and have enough land to move about freely. On the other hand, horses from smaller studs obtained higher gait scores for dressage, possibly because they are from stud farms dedicated to training horses to participate in dressage. In addition, the rider factor has a great influence on the gaits of the PRE horse, as shown in work carried out by other authors in different equine breeds, such as British sport horses (Kearsley et al., 2008). In functional studies carried out in the PRE breed (S´ anchez et al., 2014), the rider factor and the rider-horse interaction are always included in the models for estimating genetic parameters in variables related to the functionality of the PRE in dressage. A good rider tends to stabilize the horse’s locomotor pattern (Peham et al., 2004). We therefore decided to include the rider as a factor in the genetic parameter estimation model, following the indications of previous studies carried out in different equine breeds such as Anglo-Arabian (García-Ballesteros et al., 2018); Table 4 Genetic parameters for gait traits evaluated during free movement or under rider with all evaluated animals and for coincidence horses (between parentheses). σ u ±SE σ p ±SE σ r ±SE σ e ±SE h 2 ±SE r g ±SE Walk Free movement 0.18 ±0.011 (0.38 ±0.094) – – 0.73 ±0.010 (0.51 ±0.083) 0.20 ±0.011 (0.43 ±0.099) (0.49 ±0.099) Under rider 0.11 ±0.016 (0.13 ±0.034) 0.10 ±0.013 (0.05 ±0.028) 0.07 ±0.007 (0.05 ±0.009) 0.15 ±0.002 (0.34 ±0.001) 0.26 ±0.036 (0.23 ±0.047) Trot Free movement 0.26 ±0.014 (0.31 ±0.102) – – 0.90 ±0.012 (0.62 ±0.098) 0.22 ±0.011 (0.33 ±0.106) (0.61 ±0.168) Under rider 0.10 ±0.012 (0.05 ±0.018) 0.03 ±0.009 (0.05 ±0.015) 0.07±0.006 (0.04 ±0.008) 0.12 ±0.002 (0.31 ±0.001) 0.31 ±0.032 (0.11 ±0.035) Canter Free movement 0.27 ±0.016 (0.20 ±0.091) – – 0.94 ±0.014 (0.70 ±0.094) 0.22 ±0.012 (0.22 ±0.100) (0.54 ±0.193) Under rider 0.10 ±0.012 (0.04 ±0.014) 0.03 ±0.009 (0.04 ±0.013) 0.07 ±0.006 (0.02 ±0.005) 0.11 ±0.001 (0.30 ±0.001) 0.33 ±0.036 (0.10 ±0.031) σ u =Additive genetic variances, σ p =permanent environmental effect variances, σ r =rider variances, σ e =residual variances, h 2 =heritabilities, rg=genetic correlations and SE=standard error. Fig. 1. Percentage of coincident horses in the highest 25 and 10 percentile of the estimated breeding values (EBV), for each gait trait of the Pura Raza Espa˜ nola horse. M. Ripoll´ es-Lobo et al. Livestock Science 263 (2022) 105031 5 PRE (S´ anchez et al., 2014, 2017a Sol´ e et al., 2017; García-Ballesteros et al., 2018); Spanish Sport Horses (Bartolom´ e and Cockram, 2016, 2018), and Arabian horses (García-Ballesteros et al., 2018). In the last decade, numerous studies have focused on analysing functionality in horse breeds, as is the case of the Royal Dutch Sport horse (KWPN) (Ducro et al., 2009), Warmblood sport Horses (Koenen et al., 2004), Swedish Warmblood horse (Gelinder et al., 2002; Wallin et al., 2003), Menorca horses (Sol´ e et al., 2013), PRE horse (Valera et al., 2017; Sanchez et al., 2016a) and jumping horses (Ricard et al., 2020). Most studies on functional improvement in horses have focused on estimating the heritabilities of the horse’s gaits in sporting events (Olsson et al., 2000; Wallin et al., 2003; Ducro et al., 2007a, 2007b; Viklund et al., 2008; Medeiros et al., 2020). Studies that have estimated heritabilities for variables related to sport performance have produced relatively low values, which can be attributed to the high influence of environmental effects on functional traits (Barrey et al., 2002; Wallin et al., 2003), such as Hispano-Arab horses, whose heritability for the three gaits, walk, trot and canter, were 0.10, 0.13, 0.13, respectively (G´ omez et al., 2016). Considering all the animals evaluated in our study, the heritability was higher when the horse was evaluated during UR. This may be due to the fact that the UR population is smaller, with fewer animals participating, and that these animals are pre-selected by breeders to participate in this discipline. In addition, the animals participating in dressage competitions are usually related to each other, and it is very common for descendants of animals with good results in dressage to participate in these competitions. Therefore, the coancestry between them is usually greater. According to Becker et al. (2011), the values for the heritability of Oldenburger horse gaits in UR were 0.33, 0.39, and 0.34, for gait, trot, and canter, respectively. The heritability values in gaits for the PRE are slightly lower than those obtained in Oldenburger. This may be due to the greater specialization that German horses have had over time in the sport of dressage, in contrast to the PRE. However, these differences could also be attributed to the fact that the Oldengurger population analysed is more homogeneous, as the study only considered one sex (all the animals evaluated were mares). In the case of the coincidence animals in our study, the heritability was higher when the horse was evaluated during FM. The horse performed the movements in freedom and were therefore not influenced by external factors. These gaits are natural and are therefore more homogeneous and more heritable. The heritability values for FM obtained in this study were similar to the heritability of natural gaits studied in the KWPN horse by Ducro et al. (2007a), where values of 0.35 were obtained for walk, 0.50 for trot, and 0.25 for canter; in the Brazilian Horses (Medeiros et al., 2020), where values of 0.36 were obtained for walk, 0.54 for trot, and 0.35 for canter; and in Oldenburger horses, where values of 0.38 were obtained for walk, 0.49 for trot, and 0.34 for canter. In the genetic model of UR, certain external factors cannot be included, such as the type and hours of training, length of preparation of the horse, feeding, whether the rider who trains is the same rider who participates in the dressage competition, etc. The heritability values for UR were lower than those obtained in PRE competition horses (S´ anchez-Guerrero et al., 2017a), where the heritability for walk was 0.21, for trot 0.27, and 0.33 for canter. This may be due to the fact that in our study, a wider range of criteria was used to select the horses. By using criteria to select coincident animals (horses participating in FM and UR, horses ridden by more than one rider, horses having at least three dressage participations, age between 4 and 6 years), improved heritabilities were obtained. The value of genetic correlations is fundamental for building selection indices or for predicting correlated responses to selection (S´ anchez-Guerrero et al., 2017a). The genetic correlations obtained in this study between the same gaits were considerably high, with trot being the gait with the highest genetic correlation, which when analysed for FM, best reflects the trot scores in UR (Sol´ e et al., 2013). The high genetic correlations found between FM and UR gaits imply a very good predictability in dressage results for young horses. However, the correlations obtained by Becker et al. (2011) in Oldenburger horses were considerably higher, namely 0.84 (walk), 0.71 (trot), and 0.9 (canter). This may be due to the fact that in German breeds, selective efforts have achieved a significant superiority of horses in dressage competitions (Barrey et al., 2002). These results are in line with different studies in other breeds, such as Dutch Warmblood (Ducro et al., 2007a), Swedish Warmblood (Wallin et al., 2003; Viklund et al., 2008) and PRE horse (S´ anchez-Guerrero et al., 2017a), all of which have obtained positive and high genetic correlations between FM and dressage competitions. Finally, Fig. 1 shows the importance of considering the evaluation of the young horse’s natural gaits in the basic aptitude tests. In this way, breeders can to carry out an early pre-selection of PRE horses destined to participate in dressage competitions, in order to ensure a good selection response. It should be noted that, if the intensity of selection in this breed is high, the percentage of animals with the best evaluations for trot and canter, with or without rider, would be over 67%, and for walk, would be over 52%. 5. Conclusion The estimated heritability for the three natural gaits of the PRE horse, whether in free movements (FM) or under rider (UR), can be considered suitable for genetic selection of breeding stock based on genetic values. In the same way, the high genetic correlations found between the same gaits according to the type of presentation of the young horse (FM or UR) allow us to carry out an early pre-selection of animals that can participate in dressage competitions. This pre-selection entails an important economic advantage and time saving for the breeder and, indirectly, greater genetic progress in the improvement of the range of functions of the PRE horse. Author contributions Conceptualization, Valera, M; data curation and preparation, Ripoll´ es M., S´ anchez-Guerrero, M.J.; methodology, Ripoll´ es M., S´ anchez-Guerrero, M.J., Bartolom´ e, E.; formal analysis, Ripoll´ es M.; Perdomo-Gonz´ alez, D.I.; genetic parameters estimation, Ripoll´ es M., Perdomo-Gonz´ alez, D.I., results validation, Ripoll´ es M., Valera, M., Bartolom´ e, E.; discussion, Ripoll´ es M., Perdomo-Gonz´ alez, D.I., Valera, M.; writing—original draft preparation, Ripoll´ es M.; writing—review and editing, Ripoll´ es M.; supervision, Valera, M.; funding acquisition, Valera, M. All authors have read and agreed to the published version of the manuscript. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments The authors wish to thank the National Association of Pura Raza Espa˜ nol Horse Breeders (ANCCE) for providing the data used in this study. References Albertsd´ ottir, E., Eriksson, S., N¨ asholm, A., Strandberg, E., ´ Arnason, T., 2008. Genetic correlations between competition traits and traits scored at breeding field-tests in Icelandic horses. Livest Sci 114 (2–3), 181–187. Baban, M., Curik, I., Antunovic, B., Cacic, M., Korabi, N., Mijic, P., 2009. Phenotypic correlations of stride traits and body measurements in Lipizzaner stallions and mares. J. Equine Vet. Sci. 29 (6), 513–518. Barrey, E., Desliens, F., Poirel, D., Biau, S., Lemaire, S., Rivero, J.L.L., Langlois, B., 2002. Early evaluation of dressage hability in different breeds. Equine Vet. J. 34, 319–324. M. Ripoll´ es-Lobo et al. Livestock Science 263 (2022) 105031 6 Bartolom´ e, E., Cockram, M.S., 2016. Potential effects of stress on the performance of sport horses. J. Equine Vet. Sci. 40, 84–93. Bartolom´ e, E., Men´ endez-Buxadera, A., Molina, A., Valera, M., 2018. Plasticity effect of rider–horse interaction on genetic evaluations for Show Jumping discipline in sport horses. J. Anim. Breed Genet. 135 (2), 138–148. Becker, A.C., Stock, K.F., Distl, O., 2011. Genetic correlations between free movement and movement under rider in performance tests of German Warmblood horses. Livest Sci. 142, 245–252. Braam, Å., N¨ asholm, A., Roepstorff, L., Philipsson, J., 2011. Genetic variation in durability of Swedish Warmblood horses using competition results. Livest Sci. 142 (1–3), 181–187. Bruns, E., 2001. Berücksichtigung von Merkmalen der Gesundheit in der Züchtung von Sportpferden. Arch. Anim. Breed. 44 (2), 119–128. Borowska, A., Wolc, A., Szwaczkowski, T., 2011. Genetic variability of traits recorded during 100-day stationary performance test and inbreeding level in Polish warmblood stallions. Arch. Anim. Breed. 54 (4), 327–337. Brooks, S.A., Makvandi-Nejad, S., Chu, E., Allen, J.J., Streeter, C., Gu, E., McCleery, B., Murphy, B.A., Bellone, R., Sutter, N.B., 2010. Morphological variation in the horse: defining complex traits of body size and shape. Anim. Genet. 41 (2), 159–165. Bussiman, F.O., Perez, B.D.C., Ventura, R.V., Silva, F.F., Peixoto, M.G.C.D., Vizon´ a, R.G., Mattos, E.C., Sterman, J.B., Eler, J.P., Curi, R.A., Balieiro, J.C.C., 2018. Genetic analysis of morphological and functional traits in Campolina horses using Bayesian multi-trait model. Livest. Sci. 216, 119–129. Ducro, B.J., Koenen, E.P.C., Van Tartwijk, J.M.F.M., Van Arendonk, J.A.M., 2007a. Genetic relations of First Stallion Inspection traits with dressage and show-jumping performance in competition of Dutch Warmblood horses. Livest. Sci. 107 (1), 81–85. Ducro, B.J., Koenen, E.P.C., Van Tartwijk, J.M.F.M., Bovenhuis, H., 2007b. Genetic relations of movement and free-jumping traits with dressage and show-jumping performance in competition of Dutch Warmblood horses. Livest. Sci. 107 (2–3), 227–234. Ducro, B.J., Bovenhuis, H., Back, W., 2009. Heritability of foot conformation and its relationship to sports performance in a Dutch Warmblood horse population. Equine Vet. J. 41, 139–143. Galisteo, A.M., Cano, M.R., Morales, J.L., Vivo, J., Mir´ o, F., 1998. The influence of speed and height at the withers on the kinematics of sound horses at the hand-led trot. Vet. Res. Commun. 22, 415–423. García-Ballesteros, S., Varona, L., Valera, M., Guti´ errez, J.P., Cervantes, I., 2018. Crossvalidation analysis for genetic evaluation models for ranking in endurance horses. Animal 12 (1), 20–27. Gelinder, Å., Thor´ en, E., N¨ asholm, A., Philipsson, J., 2002. Relationships between test results at 3 and 4 years of age for Swedish riding horses. In: Proceedings of the 7th World Congress Genetics Applied to Livestock Production Science, Montpellier, France, 30, pp. 375–378. Giulotto, E., 2001. Will horse genetics create better champions? Trends Genet. 17 (3), 166. G´ omez, M.M., Gama, L.T., Le´ on, J.M., Fern´ andez, J., Attalla, S.A., Delgado, J.V., 2016. Genetic parameters for harmony and gaits in Hispano-Arabe horses estimated by Bayesian methods and restricted maximum likelihood. Livest Sci 188, 159–165. G´ omez-Ortiz, M.D., Molina, A., S´ anchez-Guerrero, M.J., Valera, M., 2021. Prediction of adult conformation traits from shape characteristics of Pura Raza Espa˜ nol foals. Livest. Sci 253, 104701. G´ orniak, W., Walkowicz, E., Soroko, M., Korczynski, M., 2020. Impact of the individual characteristics of French trotters on their racing performance. Turk. J. Vet. Anim. Sci. 44 (1), 110–117. Groeneveld, E. 2010. VCE user’s guide and reference manual version 6.0.ftp://ftp. tzv. fal. de/pub/vce6/doc/vce6-manual-3.1-A4. pdf. Hellsten, E.T., Viklund, Å., Koenen, E.P.C., Ricard, A., Bruns, E., Philipsson, J., 2006. Review of genetic parameters estimated at stallion and young horse performance tests and their correlations with later results in dressage and show-jumping competition. Livest. Sci 103 (1–2), 1–12. Holmstr¨ om, M., Philipsson, J., 1993. Relationships between conformation, performance and health in 4-year-old swedish warmblood riding horses. Livest Prod Sci 33, 293–312. J¨ onsson, L., N¨ asholm, A., Roepstorff, L., Egenvall, A., Dalin, G., Philipsson, J., 2014. Conformation traits and their genetic and phenotypic associations with health status in young Swedish warmblood riding horses. Livest Sci 163, 12–25. Kearsley, C.G.S., Woolliams, J.A., Coffey, M.P., Brotherstone, S., 2008. Use of competition data for genetic evaluations of eventing horses in Britain: analysis of the dressage, showjumping and cross country phases of eventing competition. Livest. Sci. 118 (1–2), 72–81. Koenen, E.P., Aldridge, L.I., Philipsson, J., 2004. An overview of breeding objectives for warmblood sport horses. Livest. Prod. Sci. 88, 77–84. Komosa, M., Frąckowiak, H., Purzyc, H., Wojnowska, M., Gramacki, A., Gramacki, J., 2013. Differences in exterior conformation between primitive, Half-bred, and Thoroughbred horses: anatomic-breeding approach. J. Anim. Sci. 91 (4), 1660–1668. Lewczuk, D., Ducro, B., 2012. Repeatability of free jumping parameters on tests of different duration. Livest. Sci. 146 (1), 22–28. Magnusson, L.E., Thafvelin, B., 1985. Relationship Between Conformation and Performance in 4-year-old Standardbred trotters. Studies on the Conformation and Related Traits of Standardbred trotters in Sweden. University of Agricultural Sciences, Swedish. Medeiros, B.R., Garbade, P., Seixas, L., Peripolli, V., McManus, C., 2020. Brazilian Sport Horse: genetic parameters for approval of Brasileiro de Hipismo stallions. Trop. Anim. Health Prod. 52 (4), 1669–1680. Molina, A., Valera, M., Galisteo, A.M., Vivo, J., G´ omez, M.D., Rodero, A., Agüera, E., 2008. Genetic parameters of biokinematic variables at walk in the Spanish Purebred (Andalusian) horse using experimental treadmill records. Livest. Sci. 116, 137–145. Olsson, E.G., ´ Arnason, T., N¨ asholm, A., Philipsson, J., 2000. Genetic parameters for traits at performance test of stallions and correlations with traits at progeny tests in Swedish warmblood horses. Livest. Prod. Sci. 65 (1–2), 81–89. Peham, C., Licka, T., Schobesberger, H., Meschan, E., 2004. Influence of the rider on the variability of the equine gait. Hum. Mov. Sci. 23 (5), 663–671. Perdomo-Gonz´ alez, D.I., G´ omez, M.D., Bartolom´ e, E., García De Paredes, R., Valera, M., 2022. Genetic parameters of morphofunctional traits in Purebred Menorca Horses based on their coat color. Animals. Submitted. Poyato-Bonilla, J., S´ anchez-Guerrero, M.J., Valera, M., Dos Santos, R.G., 2018. Population study of the Pura Raza Espa˜ nol Horse regarding its coat colour. Ann. Anim. Sci. 18 (3), 723–739. Ricard, A., Dumont Saint Priest, B., Chassier, M., Sabbagh, M., Danvy, S., 2020. Genetic consistency between gait analysis by accelerometry and evaluation scores at breeding shows for the selection of jumping competition horses. PLoS One 15 (12), e0244064. Ripoll´ es, M., S´ anchez-Guerrero, M.J., Perdomo-Gonz´ alez, D.I., Azor, P., Valera, M., 2020. Survey of risk factors and genetic characterization of ewe neck in a world population of Pura Raza Espa˜ nol Horses. Animals 10 (10), 1789. Rosengren, M.K., Sigurðard´ ottir, H., Eriksson, S., Naboulsi, R., Jouni, A., Novoa- Bravo, M., Albertsd´ ottir, E., Kristj´ ansson, P., Rhodin, M., Viklund, A., Velie, B.D., Negro, J.J., Sol´ e, M., Lindgren, G., 2021. A QTL for conformation of back and croup influences lateral gait quality in Icelandic horses. BMC Genom 22 (1), 1–13. Rustin, M., Janssens, S., Buys, N., Gengler, N., 2009. Multi-trait animal model estimation of genetic parameters for linear type and gait traits in the Belgian warmblood horse. J. Anim. Breed. Genet. 126, 378–386. S´ anchez, M.J., G´ omez, M.D., Molina, A., Valera, M., 2013a. Genetic analyses for linear conformation traits in Pura Raza Espa˜ nol horses. Livest. Sci. 157 (1), 57–64. S´ anchez, M.J., G´ omez, M.D., Pe˜ na, F., García-Monterde, J., Morales, J.L., Molina, A., Valera, M., 2013b. Relationship between conformation traits and gait characteristics in Pura Raza Espa˜ nol horses. Arch. Tierz. 56 (13), 137–148. S´ anchez, M.J., Cervantes, I., Valera, M., Guti´ errez, J.P., 2014. Modelling genetic evaluation for dressage in Pura Raza Espa˜ nol horses with focus on the rider effect. J. Anim. Breed. Genet. 131, 395–402. S´ anchez, M.J., Molina, A., G´ omez, M.D., Pe˜ na, F., Valera, M., 2016a. Relationship between morphology and performance: signature of mass-selection in Pura Raza Espa˜ nol horses. Livest. Sci. 185, 148–155. S´ anchez, M.J., Bartolom´ e, E., Valera, M., 2016b. Genetic study of stress assessed with infrared thermography during dressage competitions in the Pura Raza Espa˜ nol horse. Appl. Anim. Behav. Sci. 174, 58–65. S´ anchez-Guerrero, M.J., Cervantes, I., Molina, A., Guti´ errez, J.P., Valera, M., 2017a. Designing an early selection morphological linear traits index for dressage in the Pura Raza Espa˜ nol horse. Animal 11 (6), 948–957. S´ anchez, M.J., Azor, P.J., Molina, A., Parkin, T., Rivero, J.L.L., Valera, M., 2017b. Prevalence, risk factors and genetic parameters of cresty neck in Pura Raza Espa˜ nol horses. Equine Vet. J. 49 (2), 196–200. S´ anchez-Guerrero, M.J., Negro, S., Demyda, S., Sol´ e, M., Azor, P.J., Valera, M., 2019. Morphological and genetic diversity of Pura Raza Espa˜ nol horse with regard to the coat colour. Anim. Sci. J. 90 (1), 14–22. SAS (Statistical Analysis System, version 9.2). 2011. Available online: www.sas.com. Schroeder, W., Stock, K.F., Distl, O., 2010. Genetic evaluation of Hanoverian warmblood horses for conformation traits considering the proportion of genes of foreign breeds. Arch. Anim. Breed. 53 (4), 377–387. Sch¨ opke, K., Wensch-Dorendorf, M., Swalve, H.H., 2013. Genetic evaluations of the German Sport Horse: population structure and use of data from foal and mare inspections and performance tests of mares. Arch. Anim. Breed. 56 (1), 658–674. Sol´ e, M., G´ omez, M.D., Galisteo, A.M., Santos, R., Valera, M., 2013. Kinematic characterization of the Menorca horses at the walk and the trot: influence of hind limb pastern angle. J. Equine Vet. Sci. 33 (9), 726–732. Sol´ e, M., Santos, R., Molina, A., Galisteo, A., Valera, M., 2014. Genetic analysis of kinematic traits at the trot in Lusitano horse subpopulations with different types of training. Animal 8 (2), 192–199. Sol´ e, M., S´ anchez, M.J., Valera, M., Molina, A., Azor, P.J., S¨ olkner, J., M´ esz´ aros, G., 2017. Assessment of sportive longevity in Pura Raza Espa˜ nol dressage horses. Livest. Sci. 203, 69–75. Sol´ e, M., Valera, M., Fern´ andez, J., 2018. Genetic structure and connectivity analysis in a large domestic livestock meta-population: the case of the Pura Raza Espa˜ nol horses. J. Anim. Breed. Genet. 135, 460–471. Stewart, I.D., Woolliams, J.A., Brotherstone, S., 2011. Genetic evaluations of traits recorded in British young horse tests. Arch. Anim. Breed. 54 (5), 439–455. Valera, M., Galisteo, A.M., Molina, A., Mir´ o, F., G´ omez, M.D., Cano, M.R., Agüera, E., 2008. Genetic parameters of biokinematic variables of the trot in Spanish Purebred horses under experimental treadmill conditions. Vet. J. 178 (2), 219–226. Valera, M., 2017. La Mejora Gen´ etica Equina En La Era De Las ´ omicas: Una Historia Compartida. Universidad de Sevilla, Spain, p. 84. Pp. Vicente, A.A., Carolino, N., Ral˜ ao-Duarte, J., Gama, L.T., 2014. Selection for morphology, gaits and functional traits in Lusitano horses: I. Genetic parameter estimates. Livest. Sci. 164, 1–12. Viklund, Å., Hellsten, E.T., N¨ asholm, A., Strandberg, E., Philipsson, J., 2008. Genetic parameters for traits evaluated at field tests of 3-and 4-year-old Swedish Warmblood horses. Animal 2 (12), 1832–1841. Viklund, Å., Braam, Å., N¨ asholm, A., Strandberg, E., Philipsson, J., 2010. Genetic variation in competition traits at different ages and time periods and correlations M. Ripoll´ es-Lobo et al. Livestock Science 263 (2022) 105031 7 with traits at field tests of 4-year-old Swedish Warmblood horses. Animal 4 (5), 682–691. Viklund, Å., N¨ asholm, A., Strandberg, E., Philipsson, J., 2011. Genetic trends for performance of Swedish Warmblood horses. Livest. Sci 141 (2–3), 113–122. Viklund, A., Eriksson, S., 2018. Genetic analyses of linear profiling data on 3-year-old Swedish Warmblood horses. J. Anim. Breed Genet. 135, 62–72. Wallin, L., Strandberg, E., Philipsson, J., 2003. Genetic correlations between field test results of Swedish Warmblood Riding Horses as 4-year-olds and lifetime performance results in dressage and show-jumping. Livest. Prod. Sci. 82, 61–71. M. Ripoll´ es-Lobo et al.