COBISS: 1.08 Agris category code: L10
HERITABILITY OF LINEAR TYPE TRAITS IN THE AUTOCHTHONOUS RENDENA DUAL PURPOSE BREED
Serena MAZZA 1 2, Roberto MANTOVANI 1
ABSTRACT
The aim of the study was to estimate heritabilities (h2) of type traits and linear description scores in Rendena breed. Five traditional summary traits and 20 linear descriptions were recorded on 9,776 Rendena primiparous cows. All available pedigree information (16,234 animals) was used to set up the relationship matrix among animals. Data were analyzed using a single-trait animal model, accounting for the following effects: herd-year-classifier, days in milk, age at first calving, and the genetic additive cow effect. The most heritable trait was the stature (h2 = 0.53), whereas the lowest value was found for feet (0.13). Other considerable values of heritability were found for body size (0.45), body length (0.40), and udder width (0.43). Medium-high heritability estimates were obtained for fleshiness (0.33), and udder (0.35), and regarding linear description for thigh-buttocks traits (0.32, and 0.33), thinness and rump angle (0.36 for both traits), as well as teat length (0.33). On the other hand suspensory ligament (0.17), and thorax length (0.16) showed low heritability values. However, the general good heritability estimates suggest the possibility to use linear description or type traits in genetic evaluation of Rendena cattle.
Key words: cattle / autochthonous breeds / dual purpose / Rendena / linear type traits / heritability / Italy
1 INTRODUCTION
The Rendena breed is an autochthon dual purpose breed (milk and meat) that belongs to the "European federation of cattle breeds of the alpine system". The Rendena population is mainly raised in Trentino Alto Adige (i.e., the region of origin) and in Veneto, and cows enrolled in functional controls are 70% of total population (ANARE, personal communication). This breed exhibits small-medium size, good fertility and longevity (Mantovani et al., 1997). In Italy the breeding goal for this breed is to improve both quality and quantity of milk and meat. In the past morphological evaluation has been one of the pillars of cattle selection, because it made possible an instant evaluation of the conformation of the animals and a relatively simple classification (Forabosco et al., 2005). Genetic evaluations of type traits in Rendena cows has been implemented since 2004, but no studies have been
carried out on genetic evaluation of type traits with other productive traits. In dairy cattle, relationship of type with production (Veerkamp and Brotherstone, 1997; Kadarmideen and Wegmann, 2003) and with longevity (Larroque and Ducrocq, 2001; Vukasinovic et al., 2002) have been widely studied in order to use type traits as predictors of herd life. Studies indicated that dairy cows of moderate size, with functional udders and correct feet and legs were more likely to remain in the herd than cows without these characteristics. In beef cattle, little is known of the relationship between type traits and longevity (Forabosco et al., 2004).
As part of a wider study aimed at analysing genetic structure of type and other productive traits in the Rendena breed, the aim of this study was to estimate her-itabilities of type evaluation traits and linear description scores.
1	Dept. of Agronomy Food Natural resources Animal Environment, Agripolis, 35020 Legnaro (PD), Italy
2	Address correspondence to: serena.mazza@studenti.unipd.it
S. MAZZA and R. MANTOVANI
2 MATERIALS AND METHODS
2.1 TYPE EVALUATION
Records belonging to 11,340 primiparous Rendena cows evaluated from 1994 to 2011 were considered. Morphological evaluation consists in a mixture of different recordings: 5 traditional summary traits (i.e. type traits)
and 20 linear descriptions (Table 1). All traits are scored using 1 to 5 point scale system. Type traits include an overall score, which summarizes the general appearance of the animal and 4 different aspects of the animal (i.e. body size, fleshiness, body shape, and udder). On the other hand, linear descriptions include specific body regions regarding thorax, rump, feet, legs, thigh and udder conformation (Table 1).
Table 1: Descriptive statistics of typ	e traits, and linear description of9,776 Rendena primiparous				cows used in	the study
	Value				Descriptor	
Trait	Mean	SD	Kolmogorov-Smirnov	Min. Max.	Minimum	Maximum
Type trait						
Overall score	2.986	0.846	0.237	15	Poor	Excellent
Body size	3.115	0.866	0.257	15	Little	Large
Fleshiness	2.958	0.758	0.313	15	Poor	Excellent
Body shape	2.914	0.866	0.235	15	Fine	Heavy
Udder	3.047	0.899	0.246	15	Poor	Excellent
Linear description						
Body size						
Stature	3.127	0.975	0.230	15	Short	Tall
Body length	3.177	0.913	0.246	15	Short	Long
Thorax depth	3.186	0.880	0.254	15	Very thin	Very large
Thorax length	3.007	0.793	0.296	15	Short	Long
Fleshiness						
Shoulder, Fore view	2.790	0.833	0.278	15	Scarce	Developed
Back, Loins and Rump	2.933	0.805	0.305	15	Scarce	Developed
Thigh, Buttocks side view	3.008	0.796	0.298	15	Hollow	Rounded
Thigh, Buttocks rear view	2.848	0.814	0.278	15	Hollow	Rounded
Body shape						
Thinness	3.245	0.821	0.296	15	Heavy	Fine
Rump angle	2.669	0.680	0.368	15	Back-inclined Counter-inclined	
Rump width	3.154	0.814	0.290	15	Narrow	Broad
Rear legs side view	3.092	0.776	0.325	15	Straight	Sickle
Feet	2.885	0.709	0.338	15	Weak	Strong
Udder						
Fore udder attach	3.248	0.959	0.218	15	Loose	Tight
Rear udder attach	2.970	0.901	0.250	15	Short	Tall
Udder width	3.013	0.958	0.221	15	Narrow	Broad
Udder depth	3.362	0.716	0.321	15	Deep	Shallow
Suspensory ligament	3.216	0.828	0.319	15	Weak	Strong
Teat placement side view	2.962	0.755	0.318	15	Close	Far
Teat length	3.061	0.831	0.295	1 5	Short	Long
2.2	DATA EDITING
Original data were edited and discarded if the age at first calving was less than 24 months or greater than 60 months, if days in milk was less than 10 days or greater than 350 days, and if herds-year-classifier had less than 2 cows scored. After editing, information regarding 9,776 subjects remained for further analysis. All available pedigree information (16,234 animals) was used to set up the relationship matrix among animals.
2.3	MODELS
A preliminary ANOVA (SAS Inst. Inc., Cary, NC) was carried out to establish which non-genetic effects could be taken into account in the final model. A single trait analysis was performed using the REMLf90 program (Misztal, 2008) and applying the EM-REML algorithm.
The final model adopted for the single trait analysis was as follow:
y..., = HYC + DIM. + AFC + u, + e..M, 1 ljkl i j k l ljkl'
where yijkl is the type score or linear description for cow l, HYC is the fixed effect of the herd-year-classifier i (1,380 different levels), DIM is the fixed effect of days in milk j (8 classes from 10 to 30 day after calving and from 31 to 210 day after calving using 30-day intervals, or for later evaluation >210 d), AFCk is the fixed effect of age at first calving k (8 classes: < 26 months, from 27 to 38 using 2-month intervals, and > 39 months for the last class),
u is the random additive effect of cow l, and e is the
l ' ijkl
random residual term.
Due to program limitations, the SE for estimated heritabilities were obtained using the following formula (Falconer, 1989):
where t is intraclass correlation obtained by (h2/4) for paternal half-sib estimates, k is the average number of offspring per sire, and s is the number of sires.
<	0.001). Moreover, DIM and AFC were significant for almost all traits (P < 0.001), except for the rump angle, feet and teat placement side view for the first factor, and body shape, rump angle and rear legs side view for the second factor (Table 2). The lower significant values were found for udder (P < 0.01), and rear legs side view (P
<	0.05) for DIM, and udder depth (P < 0.05) for AFC. The coefficient of determination (R2) ranged between 0.19 (e.g., thinness) and 0.36 (e.g., shoulder - fore view). Moreover, for all fleshiness traits (e.g, shoulder, back, loins and rump, thigh and buttocks), determination coefficients are greater than 0.30 (Table 2).
3.1 HERITABILITIES
Estimated variance and heritability values are shown in Table 3. Analyzing the results for all morphological traits, the most heritable trait was the stature (0.53), whereas the lowest value of heritability was found for feet (0.13). Moreover the heritability for overall score was quite high (0.26), and also the other four type traits exhibited appreciable values, with the exception of body shape that showed the smaller value (0.17). Regarding linear descriptions, thorax length also showed a low value (0.16). On the other, hand all fleshiness traits had a rather high heritability, from 0.29 for shoulder -fore view, to 0.33 for thigh - buttocks side view. Significantly lower heritability values for body size, fleshiness and thinness were reported for Piemontese hypertrophic cows by Mantovani et al., 2010 (i.e., 0.26, 0.15 and 0.12 respectively).
Among the seven udder traits only the suspensory ligament exhibited a low heritability value (0.17), while the others ranged from 0.30 to 0.43. Similar value of heritability was reported by Wall et al. (2005) for suspensory ligament in a study conducted on Holstein-Friesian cows (i.e., 0.16). Slightly greater heritability values for udder depth and for teat placement side view were reported for Holstein-Friesian primiparous cows by Berry et al., 2004 (i.e., 0.33 and 0.38).
Standard errors of heritabilities resulted generally low, ranging from 0.022 to 0.038. This may be attributed to the larger data set used, derived from measurements obtained over 18 years.
3 RESULTS AND DISCUSSION
With only few exceptions, all analyzed non genetic fixed factors reported a significant amount of the total variance when analyzed with ANOVA (Table 2). The HYC effect was significant for all the traits considered (P
4 CONCLUSIONS
Results of this study indicate that herd-year-classifier, days in milk and age at first calving were the most significant effects and they most affect linear type traits in Rendena breed. Furthermore estimates of heritabilities
Table 2: Results of ANOVA on nongenetic fixed factors used in the final model using type traits, and linear description of9,776 Ren-dena primiparous cows
Mean squares
Trait
HYC1
DIM2
AFC3
Residual
Type trait Overall score Body size Fleshiness Body shape Udder Linear description Body size Stature Body length Thorax depth Thorax length Fleshiness
Shoulder, Fore view Back, Loins and Rump Thigh, Buttocks side view Thigh, Buttocks rear view Body shape Thinness Rump angle Rump width Rear legs side view Feet Udder
Fore udder attach Rear udder attach Udder width Udder depth Suspensory ligament Teat placement side view Teat length
1.95*** 2.13*** 2.84*** 1.75*** 1.72***
2.00*** 2.08*** 2.07*** 2.63***
3.37*** 2.99*** 2.67*** 2.77***
1.38*** 1.76*** 2.40*** 1.74*** 1.76***
2.22*** 1.66*** 1.73*** 1.91*** 1.62*** 1.81*** 1.54***
3.50*** 7.21*** 16.93*** 1.60 14.36***
5.68*** 6.46*** 4.17*** 16.28***
12.98*** 22.31*** 16.05*** 15.87***
3.42*** 1.95 12.33*** 0.89 4.37***
17.34*** 12.52*** 8.06*** 2.16* 16.58*** 4.14*** 5.82***
8.82*** 27.29*** 19.81*** 6.14*** 2.70**
8.72*** 36.65*** 45.50*** 34.51***
13.43*** 14.32*** 18.37*** 26.44***
4.03*** 1.01 83.17*** 2.26* 2.00
9.08*** 3.64*** 5.49*** 18.45*** 18.60*** 1.28 5.42***
0.79 0.79 0.67 0.82 0.85
0.91 0.84 0.81 0.71
0.72 0.71 0.71 0.72
0.80 0.65 0.73 0.74 0.67
0.87 0.85 0.91 0.67 0.78 0.71 0.80
0.25 0.28 0.33 0.23 0.23
0.26 0.28 0.28 0.32
0.36 0.34 0.32 0.33
0.19 0.23 0.32 0.23 0.23
0.29 0.23 0.23 0.24 0.23 0.24 0.22
1	HYC = effect of herd-year-classifier (i.e., 1,380 different levels);
2	DIM = effect of days in milk (8 classes: from 10 to 30 day after calving and from 31 to 350
3	AFC = effect of age class at first calving (8 classes: from 24 to 26 month and from 27 to 60 * P < 0.05, ** P < 0.01, *** P < 0.001
day using 30-day intervals); month using 2 mo intervals);
obtained in this research indicate that body size is the most hereditable trait, with the except of thorax length. However all analyzed traits showed good heritability values considering that data were obtained from field condition. These rather high heritability indicate a good
possibility for using linear description or type traits in routinely genetic evaluation of Rendena cattle.
R
Table 3: Estimated variance and heritabilities for type traits, and linear description of Rendena primiparous cows
Parameter1
Trait	a2 g	a2 e	h2	SE h
Type trait				
Overall score	0.165	0.476	0.26	0.027
Body size	0.298	0.369	0.45	0.035
Fleshiness	0.153	0.311	0.33	0.030
Body shape	0.118	0.564	0.17	0.023
Udder	0.262	0.491	0.35	0.031
Linear description				
Body size				
Stature	0.468	0.412	0.53	0.038
Body length	0.298	0.439	0.40	0.033
Thorax depth	0.195	0.478	0.29	0.028
Thorax length	0.081	0.426	0.16	0.024
Fleshiness				
Shoulder, Fore view	0.151	0.377	0.29	0.028
Back, Loins and Rump	0.155	0.361	0.30	0.029
Thigh, Buttocks side view	0.171	0.350	0.33	0.030
Thigh, Buttocks rear view	0.174	0.362	0.32	0.030
Body shape				
Thinness	0.236	0.425	0.36	0.031
Rump angle	0.154	0.270	0.36	0.032
Rump width	0.133	0.411	0.24	0.027
Rear legs side view	0.115	0.438	0.21	0.025
Feet	0.059	0.396	0.13	0.022
Udder				
Fore udder attach	0.238	0.552	0.30	0.029
Rear udder attach	0.234	0.523	0.31	0.029
Udder width	0.374	0.498	0.43	0.034
Udder depth	0.141	0.327	0.30	0.029
Suspensory ligament	0.105	0.518	0.17	0.023
Teat placement side view	0.160	0.362	0.31	0.029
Teat length	0.219	0.436	0.33	0.030
1 a2g is the additive genetic variance, a2e is the random residual variance, h2 is the estimated heritability, and h2 is the SE of the estimated heritability.
5 REFERENCES
Berry D.P., Buckley F., Dillon P., Evans R.D., Veerkamp R.F. 2004. Genetic relationship among linear type traits, milk yield, body weight, fertility and somatic cell count in primiparous dairy cows. Irish Journal of Agriculture and Food Research, 43: 161-176 Falconer D.S. 1989. Quantitative genetics. 3rd ed. Harlow, UK
Longman Scientific and Technical. Forabosco F., Groen A.F., Bozzi R., Van Arendonk J.A.M., Filip-pini F., Boettcher P., Bijma P. 2004. Phenotypic relationship between longevity, type traits, and production in Chianina beef cattle. Journal of Animal Science, 82: 1572-1580 Forabosco F., Rossetti E., Sarti F.M., Panella F., Bittante G., Mantovani R., Filippini F. 2005. Morphological factorials for females of the Chianina, Marchigiana and Romagnola breeds. In: 4th World Italian Beef Cattle Congress, Gubbio, 29 Apr.-1 May 2005 ANABIC: 167-168 Kadarmideen H.N., Wegmann S. 2003. Genetic parameters for body condition score and its relationship with type and production traits in Swiss Holstein. Journal of Dairy Science, 86: 3685-3693 Larroque H., Ducrocq V. 2001. Relationship between type and longevity in the Holstein breed. Genetic Selection and Evolution, 33: 39-59 Mantovani R., Cassandro M., Contiero B., Albera A., Bittante G. 2010. Genetic evaluation of type traits in hypertrophic Piemontese cows. Journal of Animal Science, 88: 35043512
Mantovani R., Gallo L., Carnier P., Cassandro M., Bittante G. 1997. The use of a juvenile selection scheme for genetic improvement of small population: the example of Rendena breed. In: 48th Annual Meeting of the European Association of Animal Production, Vienna, 25-28 Aug. 1997 (unpublished)
Misztal I. 2008. Reliable computing in estimation of variance components. Journal of Animal Breeding and Genetics, 125: 363-370
Veerkamp R.F., Brotherstone S. 1997. Genetic correlations between linear type traits, food intake, live weight and condition score in Holstein-Friesian dairy cattle. Animal Science, 64: 385-392
Vukasinovic N., Schleppi Y., Kùnzi N.M. 2002. Using conformation traits to improve reliability of genetic evaluation for herd life based on survival analysis. Journal of Dairy Science, 85: 1556-1562 Wall E., White I.M.S., Coffey M.P., Brotherstone S. 2005. The relationship between fertility, rump angle, and selected type information in Holstein-Friesian cows. Journal of Dairy Science, 88: 1521-1528