Breed and heterosis effects for milk yield traits at different production levels, lactation number and milking frequencies

ABSTRACT

The objective of this study was to evaluate the effects of production level (PL) and lactation number (LN) on the expression of breed, and heterosis for yield traits in cows milked once (OAD) or twice daily (TAD) in New Zealand. Records were used from 35,192 Friesian (F), 31,118 Jersey (J) and 88,606 crossbred (F × J) cows that calved in spring between 2008 and 2012. With an average LN of five, F cows had a higher milk yield (592 kg in OAD and 909 kg in TAD), fat yield (2.1 kg in OAD and 6.3 kg in TAD) and protein yield (11.9 kg in OAD and 19.5 kg in TAD) compared with J cows. Expressed as a percentage, heterosis effects for milk yield traits ranged from 4.1%–7.6% for production traits, and were similar across milking frequencies. In absolute values, however, heterosis was different for fat and protein yield only in first-lactation cows, and no differences were found in milk yield. In a second analysis, breed effects, defined as F–J, increased as PL of the herd increased in both OAD and TAD systems. The highest expressions of heterosis were found at medium (5.9%–6.4%) and low (5.2%–6.8%) PL in cows milked OAD and TAD, respectively. However, in absolute values, the greatest heterosis was observed at high PL in the population milked TAD. The estimates of breed and heterosis obtained in this study can be used for simulation studies to evaluate the profitability of crossbreeding systems under OAD and TAD milking systems in New Zealand dairy farms, at different PL.

KEYWORDS:

• Breed effects
• dairy cattle
• heterosis effects
• milking frequency
• New Zealand
• production level

Introduction

Traditionally, cows in milk production systems in New Zealand have been milked twice a day (TAD). However, there has been increasing popularity of milking once a day (OAD) for herd management and lifestyle benefits (Davis et al. 1999; Clark et al. 2006).

Crossbreeding between New Zealand Holstein-Friesian (F) and Jersey (J) has been adopted since the 1960s to upgrade J into F cows. However, since 1985, crossbreeding has been implemented as a mating strategy (Montgomerie 2005). This has brought favourable heterosis for production, fertility and survival traits, resulting in increased overall farm profitability (López-Villalobos et al. 2000). For New Zealand’s dairy cattle, Harris (2005) summarised heterosis effects for economically important traits under TAD milking, but so far, no studies have quantified breed and heterosis effects in cows milked OAD.

Several studies (Bryant et al. 2007; Penasa et al. 2010a; Kargo et al. 2012) have evaluated dairy cattle performance and expression of heterosis effects at different production levels (PL). In those studies, breed and heterosis effects have varied across PL.

Crossbreeding in animal production systems has been considered beneficial when management (environmental–nutritional) conditions are poor (Kargo et al. 2012). In an extensive review, Barlow (1981) concluded that heterosis is better expressed when the environmental conditions are suboptimal. Nonetheless, Bryant et al. (2007) found low or no heterosis in restricted environments, under TAD systems, in New Zealand dairy cattle.

The objective of this study was to evaluate the effects of PL and lactation number (LN) on the expression of breed and heterosis effects for milk yield traits in cows milked OAD or TAD in New Zealand.

Materials and methods

Data

Lactation records for milk yield (MY), fat yield (FY) and protein yield (PY) recorded from 2008 to 2012, and pedigree information, were provided by Livestock Improvement Corporation (Hamilton, New Zealand). Data were edited as follows:

1. Lactation records were sorted based on codes indicating whether the cow was milked OAD or TAD. Herds milked OAD were those in which 100% of the cows were milked OAD. Using GPS Visualizer (Schneider 2012), TAD herds were selected within a radius of 20 km of OAD herds. In some cases, in a given single map coordinate, an OAD farm was surrounded by several TAD herds; in such cases, all TAD herds were selected.

2. Herds with fewer than 50 cows were removed from the data set.

3. Only records from spring calving cows in their first five lactations with lactation lengths between 150 and 305 days were considered.

4. Only records from F, J and their crosses (F × J) were included, discarding cows of unknown breed and pedigree.

The breed composition of each cow was calculated using the following equation:where is the proportion of genes from breeds F or J in the progeny, and and are the proportions of breeds F or J in the sire and dam, respectively. Pure breed cows were considered when their breed composition was ≥93.75% from a particular breed.

Coefficient of heterosis in a cow was calculated using the following equation (Dickerson 1973):where and are the proportions of F or J in the sire and and are the proportions of J and F in the dam, respectively.

The final data set used in this analysis contained 124,620 lactation records from 298 herds milked OAD and 194,631 records from 350 herds milked TAD. The population studied included 9122 (5.8%) and 26,239 (16.8%) pure F cows milked OAD and TAD, respectively. The pure J cows were: 18,417 (11.8%) and 13,129 (8.4%) milked OAD and TAD, respectively. Finally, F × J cows milked OAD were 38,180 (24.5%) and 50,956 (32.7%) milked TAD.

Three groups per milking frequency (MF) were constructed based on herd PL (low, medium or high) for milk solids (fat + protein) per cow using the FASTCLUS procedure of SAS version 9.3 (SAS Institute Inc). The procedure uses Euclidean distances, guaranteeing that the distances between all the observations in the same cluster are less than the distances between observations in different clusters. Number of herds and average milk solids per PL are presented in Table 1.

Table 1. Number of herds and lactations, and mean of milk solids (kg/cow) by production level and milking frequency.

Milking frequency	Production level	Number of herds	Number of lactations	Milk solids^a (kg/cow)
OAD	Low	110	30,126	203.6
	Medium	141	60,165	269.2
	High	47	34,329	339.9
TAD	Low	168	62,960	272.7
	Medium	150	86,761	353.8
	High	32	44,910	434.1

^aDefined as the sum of fat and protein yields.

Statistical analysis

A univariate linear model was used to obtain breed and heterosis effects for total lactation yield of milk, fat and protein using the ASREML 3.0 software package (Gilmour et al. 2009). The model included the fixed effects of MF, LN, interaction between MF and LN; linear regression of MY, FY or PY on mean calving date deviation from median calving date of the herd for a given season; linear regression of MY, FY or PY on mean days in milk deviation from median days in milk of the herd for a given season; linear regressions of MY, FY or PY on gene proportion of F per MF; linear regressions of MY, FY or PY on coefficient of expected heterosis per MF; and the random effects of herd–season, additive genetic effect of cow, permanent environmental effect of cow and residual error.

For each MF, the estimate of the regression coefficient of the proportion of F and linear effect of heterosis, gives, respectively, the estimation of the breed effect between F and J and heterosis in the first cross cow (Back & López-Villalobos 2007).

A second univariate linear model was implemented to evaluate the effect of PL. The model was similar to the model described above, but included the fixed effect of PL and the interaction between MF and PL.

Results

Number of herds and average milk solids per PL are presented in Table 1. Tables 2 and 3 present predicted performances of pure F and J and first crosses (F₁) F × J, and breed and heterosis effects for MY, FY and PY for each MF by LN and PL, respectively.

Table 2. Predicted means and SEM of production traits for Holstein-Friesian (F), Jersey (J) and first cross (F₁) F × J cows and estimates of breed and heterosis effects by milking frequency and lactation number.

Lactation number	MF	F	F1F × J	J	Breed effect	Heterosis effect
					F–J (kg)	kg*	%**
Milk yield (kg/cow)
1	OAD	2402 ± 26	2343 ± 25	2075 ± 26	327^a ± 14	105^a ± 11	4.7
1	TAD	3345 ± 21	3224 ± 21	2822 ± 22	523^b ± 12	141^a ± 9	4.6
2	OAD	3105 ± 26	3024 ± 25	2610 ± 26	495^a ± 13	167^a ± 11	5.8
2	TAD	4063 ± 21	3827 ± 21	3267 ± 22	796^b ± 12	162^a ± 9	4.4
3	OAD	3486 ± 26	3349 ± 25	2868 ± 26	618^a ± 13	172^a ± 11	5.4
3	TAD	4520 ± 21	4203 ± 21	3547 ± 22	973^b ± 12	170^a ± 9	4.2
4	OAD	3689 ± 26	3552 ± 25	2978 ± 26	711^a ± 14	219^a ± 12	6.6
4	TAD	4741 ± 21	4360 ± 21	3637 ± 22	1104^b ± 12	171^b ± 10	4.1
5	OAD	3813 ± 27	3605 ± 25	3004 ± 26	809^a ± 15	197^a ± 14	5.8
5	TAD	4821 ± 22	4441 ± 21	3670 ± 22	1151^b ± 13	196^a ± 11	4.6
Fat yield (kg/cow)
1	OAD	117.7 ± 1.3	122.5 ± 1.2	115.6 ± 1.3	2.1^a ± 0.7	5.8^a ± 0.6	5.0
1	TAD	152.0 ± 1.0	162.4 ± 1.0	151.1 ± 1.1	0.9^a ± 0.6	10.9^b ± 0.5	7.2
2	OAD	147.9 ± 1.3	156.4 ± 1.2	147.3 ± 1.3	0.6^a ± 0.7	8.8^a ± 0.6	6.0
2	TAD	182.9 ± 1.0	193.4 ± 1.0	179.5 ± 1.1	3.4^b ± 0.6	12.2^b ± 0.5	6.7
3	OAD	163.8 ± 1.3	173.6 ± 1.2	163.0 ± 1.3	0.8^a ± 0.7	10.2^a ± 0.6	6.2
3	TAD	202.7 ± 1.0	211.7 ± 1.0	196.5 ± 1.1	6.2^b ± 0.6	12.1^a ± 0.5	6.1
4	OAD	171.5 ± 1.3	183.5 ± 1.2	169.5 ± 1.3	2.0^a ± 0.7	13.0^a ± 0.7	7.6
4	TAD	212.2 ± 1.0	220.7 ± 1.0	202.4 ± 1.1	9.8^b ± 0.6	13.4^a ± 0.5	6.5
5	OAD	176.2 ± 1.3	185.5 ± 1.2	171.2 ± 1.3	5.0^a ± 0.8	11.8^a ± 0.8	6.8
5	TAD	215.8 ± 1.1	225.0 ± 1.0	204.4 ± 1.1	11.4^b ± 0.7	14.9^b ± 0.6	7.1
Protein yield (kg/cow)
1	OAD	91.7 ± 1.0	92.4 ± 1.0	84.4 ± 1.0	7.3^a ± 0.5	4.3^a ± 0.4	4.9
1	TAD	121.4 ± 0.8	122.4 ± 0.8	110.2 ± 0.8	11.2^b ± 0.5	6.6^b ± 0.3	5.7
2	OAD	118.6 ± 1.0	121.0 ± 1.0	108.9 ± 1.0	9.7^a ± 0.5	7.3^a ± 0.4	6.4
2	TAD	148.6 ± 0.8	148.0 ± 0.8	131.7 ± 0.8	16.9^b ± 0.4	7.9^a ± 0.4	5.6
3	OAD	132.6 ± 1.0	134.4 ± 1.0	120.6 ± 1.0	12.0^a ± 0.5	7.8^a ± 0.4	6.2
3	TAD	164.5 ± 0.8	162.3 ± 0.8	144.0 ± 0.8	20.5^b ± 0.4	8.1^a ± 0.4	5.2
4	OAD	139.2 ± 1.0	142.0 ± 1.0	125.4 ± 1.0	13.8^a ± 0.5	9.7^a ± 0.5	7.3
4	TAD	171.8 ± 0.8	168.3 ± 0.8	148.0 ± 0.9	23.8^b ± 0.5	8.4^a ± 0.4	5.3
5	OAD	143.2 ± 1.0	143.4 ± 1.0	126.3 ± 1.0	16.9^a ± 0.5	8.7^a ± 0.5	6.4
5	TAD	174.2 ± 0.8	171.1 ± 0.8	149.3 ± 0.9	24.9^b ± 0.5	9.3^a ± 0.4	5.8

SEM, standard error of the mean.

MF, milking frequency; OAD, milking once daily; TAD, milking twice daily.

*Expressed as F₁ F × J − ([F + J]/2).

**Expressed as a percentage of heterosis effects relative to the phenotypic average of the parental breeds under milking frequency and lactation number, as appropriate.

^a,bWithin traits and production level, breed and heterosis effects with different superscripts were significantly different between milking frequencies (P < 0.05).

Table 3. Predicted means and SEM of production traits for Holstein-Friesian (F), Jersey (J) and first cross (F₁) F × J cows and estimates of breed and heterosis effects at different production levels.

Production level	MF	F	F1 F × J	J	Breed effect	Heterosis effect
					F–J (kg)	kg*	%**
Milk yield (kg/cow)
L	OAD	2554 ± 25	2499 ± 23	2178 ± 25	376^a ± 18	133^a ± 14	5.6
	TAD	3529 ± 17	3351 ± 16	2843 ± 18	686^b ± 13	165^a ± 10	5.2
M	OAD	3313 ± 20	3221 ± 19	2770 ± 20	543^a ± 13	179^a ± 10	5.9
	TAD	4471 ± 17	4174 ± 16	3577 ± 18	894^b ± 12	150^a ± 9	3.7
H	OAD	4184 ± 28	3903 ± 27	3369 ± 17	815^a ± 16	126^a ± 13	3.3
	TAD	5548 ± 26	5139 ± 26	4247 ± 28	1301^b ± 16	241^b ± 13	4.9
Fat yield (kg/cow)
L	OAD	121.3 ± 1.2	128.7 ± 1.1	120.2 ± 1.2	1.1^a ± 0.9	7.9^a ± 0.7	6.5
	TAD	158.8 ± 0.8	167.3 ± 0.8	154.6 ± 0.9	4.2^b ± 0.6	10.6^b ± 0.5	6.8
M	OAD	157.7 ± 1.0	167.1 ± 0.9	156.5 ± 0.9	1.2^a ± 0.6	10.0^a ± 0.5	6.4
	TAD	202.8 ± 0.8	212.4 ± 0.8	198.1 ± 0.8	4.7^b ± 0.6	11.9^a ± 0.4	5.9
H	OAD	195.5 ± 1.3	203.1 ± 1.3	192.0 ± 1.3	3.5^a ± 0.8	9.3^a ± 0.6	4.8
	TAD	246.8 ± 1.2	258.4 ± 1.2	238.1 ± 1.3	8.7^b ± 0.8	15.9^b ± 0.6	6.6
Protein yield (kg/cow)
L	OAD	97.3 ± 1.0	99.0 ± 0.9	88.9 ± 0.9	8.4^a ± 0.6	5.9^a ± 0.5	6.3
	TAD	127.7 ± 0.7	127.6 ± 0.6	113.1 ± 0.7	14.6^b ± 0.5	7.2^a ± 0.3	6.0
M	OAD	126.4 ± 0.8	128.6 ± 0.7	115.3 ± 0.7	11.1^a ± 0.5	7.7^a ± 0.4	6.4
	TAD	163.2 ± 0.6	161.4 ± 0.6	144.1 ± 0.7	19.1^b ± 0.4	7.7^a ± 0.3	5.0
H	OAD	158.9 ± 1.1	156.8 ± 1.0	141.5 ± 1.1	17.4^a ± 0.5	6.6^a ± 0.4	4.4
	TAD	203.2 ± 1.0	199.0 ± 1.0	173.9 ± 1.0	29.3^b ± 0.6	10.4^b ± 0.4	5.5

SEM, standard error of the mean.

H, high milk solids yield; L, low milk solids (fat + protein) yield; M, medium milk solids yield; MF, milking frequency; OAD, milking once daily; TAD, milking twice daily.

*Expressed as F₁ F × J − ([F + J]/2).

**Expressed as a percentage of heterosis effects relative to the phenotypic average of the parental breeds under milking frequency and production levels, as appropriate.

^a,bWithin traits and production level, breed and heterosis effects with different superscripts were significantly different between milking frequencies (P < 0.05).

Breed effects—defined as the difference between production performances of F and J—indicate that, overall, F cows were superior for the production traits compared with J cows in both OAD and TAD systems. However, the superiority of F cows was higher in TAD. The breed effect increased in successive lactations in all of the three traits studied, with the exception of FY under OAD systems, where the highest breed effect was found in fifth-lactation cows followed by first-lactation cows. Estimated yields (average of the five LN) of F compared with J cows were 317 kg of milk, 4.2 kg of fat and 7.5 kg of protein greater in TAD compared with OAD (Table 2).

In mature cows, heterosis effects across MF were 5.4%–7.6% in OAD and 4.1%–7.1% in TAD systems. In first-lactation cows milked OAD, however, the relative values of heterosis ranged from 4.7%–5.0%; different from the 4.6%–7.2% found in first-lactation cows milked TAD (Table 2).

The findings show that estimates of heterosis effects were not consistently greater in TAD compared with OAD. Compared with OAD systems, estimates of heterosis were similar in MY, except in fourth-lactation cows. For FY and PY, estimates of heterosis effects were greater in first-lactation cows; further second- and fifth-lactation cows milked TAD had significantly greater heterosis for FY (Table 2).

In general, breed effects increased in favour of F as PL of the herds increased in both OAD and TAD systems. The smaller breed difference at low and medium PL compared with high PL suggest that J cows might have an advantage over F cows in those environments, in particular in OAD systems (Table 3).

Heterosis effects, expressed in relative values at low PL were slightly greater compared with high PL in TAD systems, and the lowest percentage of heterosis effects was observed at medium PL. In absolute values, however, heterosis estimates were greatest at high PL in the population milked TAD. In OAD systems, the relative heterosis effects for MY, FY and PY were greater at low and medium compared with high PL. Percentage of heterosis at low and medium PL ranged between 5.6%–6.5%, and at high PL ranged between 3.3%–4.8% (Table 3).

Discussion

The results presented in Table 2 show greater breed effects in TAD than in OAD, suggesting, in agreement with Clark et al. (2006), a breed × MF interaction.

Generally, cows under OAD systems have reduced availability of dry matter due to higher stocking rates (Cooper & Clark 2001). The smaller breed differences for the production traits in OAD milking suggest, in consequence, that J cows are better adapted than F cows to OAD systems.

Similar to the results of the present study, several researchers (Macmillan et al. 1981; Ahlborn-Breier & Hohenboken 1991; Bryant et al. 2007; Penasa et al. 2010b) observed superiority in production traits of F compared with J in grazing conditions. The results confirm the greater superiority of F cows regarding milk traits. However, in dairy systems where the payment scheme gives greater emphasis to milk solids, as is the case of New Zealand, the J breed has an important role in the industry. For instance in New Zealand, F and J cows have similar milk value per lactation when milk processing is 100% whole milk powder and skim milk powder; and J cows have greater milk value per litre in a payment scheme that favours the production of fat and protein, and penalises volume (Sneddon et al. 2015). In addition, smaller sized cows such as J, compared with F, are often preferred under grazing conditions (Prendiville et al. 2009, 2011).

In this study, the percentage of heterosis found for productive traits is similar to the 5.5% reported by Prendiville et al. (2010) in F and J cows under TAD milking under Irish grazing conditions. In New Zealand, Ahlborn-Breier & Hohenboken (1991) had reported for first-lactation first-cross cows milked TAD, a heterosis of 6.1% and 7.1% for MY and FY, respectively. These percentages agree with the results presented in Table 2 for FY, but are greater for MY in cows milked TAD.

In absolute values, however, Harris (2005) reported heterosis lower than our findings (in both MF) (139, 7.7 and 5.5 kg of MY, FY and PY, respectively) in New Zealand under TAD systems. Comparing that study with values presented in Table 3, heterosis effects are similar only for the milk traits at low PL herds milked OAD. Likely differences between production performances of the dairy cattle populations might explain this discrepancy.

The heterosis effects found in TAD systems in this study are in agreement with research carried out in Europe in North American Holstein-Friesian crossbred with Dutch-Friesian (Van der Werf & de Boer 1989), Black and White cattle (Boichard et al. 1993) and European Friesian (Akbas et al. 1993), but lower than the results found by Penasa et al. (2010b) in Ireland with crosses of North American Holstein-Friesian with F, J and Montbéliarde.

Across MF, the estimates of heterosis were similar in mature cows milked OAD and TAD. It is possible that the greater reduction in first-lactation cows milked OAD, compared with the group milked TAD, is in part due to the limitation in the ability for milk storage (Clark et al. 2006) and might indicate this interaction. Consequently, both breeds F and J might not only express their additive genetic effects for milk production under OAD systems, but also the favourable dominance and epistatic effects of their crosses. The interaction between MF and LN also could be due to first-lactation cows being less capable of competing for forage, consequently, the low dry matter availability under OAD milking may have reduced the expression of heterosis in first-lactation cows milked OAD compared with the corresponding group milked TAD.

Estimates of heterosis were lower in first-lactation cows than in mature cows under both milking systems. McAllister (1986) suggested that this might be due to different gene expression, either intra-locus or inter-locus, affecting LN. McAllister reported lower heterosis for MY and FY in third-lactation cows (0.7%–0.8%), compared with first- (3.7%–3.9%), second- (1.5%–3.4%) and fourth-lactation cows (1.6%–2.6%) in Holstein-Friesian and Ayrshire hybrids (F₁). These findings differ from the results presented in Table 2; however, comparison of purebreds and crossbreds must consider records over several LN in order to evaluate profitability of crossbreeding in the production system.

The superiority of F in TAD systems and in high PL indicates a genotype × environment interaction. In particular, the genotype × environment interaction found in this study can be referred to as scaling effect, which is when phenotypic performance between breeds increases accordingly with more favourable environments (Bryant et al. 2005). Studies by Bryant et al. (2007), Penasa et al. (2010a) and Kargo et al. (2012) indicate that in general, more productive cows (with a large proportion of North American genes) increased their superiority in higher input systems compared with F, J and Dutch-Friesian. In more intensive systems, the nutritional requirements of highly productive cows are better met (Penasa et al. 2010a) allowing high-producing cows to more fully express their genetic merit for milk production.

The amount of FY and PY of F cows milked OAD in low and medium PL was considerably smaller compared with high PL, suggesting that J cows might have an advantage over F cows in those environments. The nutritional status of cows in grazing conditions varies considerably across the seasons; hence, F cows cannot express their potential when intake is restricted (Ahlborn-Breier & Hohenboken 1991).

According to Barlow (1981), estimates of heterosis tend to be smaller in less favourable environments, which are supported by the results obtained by Panesa et al. (2010a). However, the latter study refers to a crossbred population upgrading from Dutch-Friesian into North American Holstein-Friesian, which might have generated a bias.

Studies by Bryant et al. (2007) and Kargo et al. (2012) reported heterosis × environment interaction for milk traits ranging from 2.7%–9.5%. The results presented in this study are into that range. In absolute values, greater heterosis was found in high PL milked TAD, indicating a scaling effect on the expression of heterosis. The average PL in the present study, and in Bryant et al. (2007) (with crosses of F and J with overseas Holstein-Friesian) and in Kargo et al. (2012) between two strains of J (Danish and North American) were considerably lower than in the study of Penasa et al. (2010a), indicating that the evaluation of heterosis expression must consider the environment and the breeds involved.

The results presented in this study are important because farmers generally cull the less-productive cows on the basis on their production worth (PW) index, which represents the genetic superiority or inferiority of a cow to convert 5 t of feed dry matter into farm profit (Dairy NZ 2015). This index allows farmers to compare cows of different breeds and ages over a typical lifetime (Montgomerie 2005). Production worth considers the production values for MY, PY, FY and live weight, each weighted by their respective economic values (Holmes et al. 2002). Production values are calculated as the sum of estimated breeding values plus heterosis effects and permanent environmental effects (Holmes et al. 2002). Therefore, PW is higher in crossbred F × J (and their backcrosses) than in pure breeds (Montgomerie 2005). For instance, López-Villalobos et al. (2000) with a deterministic model under pastoral conditions in New Zealand showed that rotational crossbreeding systems were more profitable (net income per hectare) than milk production with pure breeds. Under OAD conditions, PW of crossbred cows might be relatively higher than their counterparts milked TAD, because the benefits of crossing animals are generally achieved when the genetic differences between purebreds are low (Falconer & Mackay 1996). The smaller breed differences between F and J suggest that relative to pure breeds, crossbreeding in OAD systems could increase farm profitability by a greater magnitude than in TAD.

Conclusions

Breed effects for production traits between F and J cows in New Zealand differed across LN and MF. Breed performance is strongly influenced by PL. The results suggest that F cows are better suited to high PL environments milked TAD, in contrast with J cows which are more adapted to low–medium PL environments milked OAD.

Lactation number and PL are factors that affect the expression of heterosis. The estimates of breed and heterosis reported in this study can be used for simulation studies to evaluate if crossbred cows under OAD systems are more profitable than F and J cows, as found in TAD milking systems under New Zealand grazing conditions at different production levels.

Acknowledgements

The principal author acknowledges the support from Programa Formación de Capital Humano Avanzado, Becas Chile, CONICYT doctoral scholarship and information provided by Livestock Improvement Corporation (Hamilton, New Zealand).

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

N López-Villalobos http://orcid.org/0000-0001-6611-907X