Dairy cattle genetic evaluation in EU countries

Mohammed Sallam, Joanna Sendecka, Valentina Palucci and Alexis Michenet

Interbull Center, Uppsala-Sweden

Twenty-two EU member countries have established the systems for dairy cattle genetic evaluations. These countries, in addition to other outside Europe countries, are collecting within-country performance records for national evaluation. The results from national evaluations are shared with Interbull Center and processed by Interbull geneticist in International (across countries) evaluation. Not only results of national evaluations that are shared with Interbull but also information/descriptions about each national evaluation system including how performance data have been collected, adjusted, analyzed, validated, published etc. However, this information still provided in different expressions, different form, incomplete and thus difficult to have an overview between countries and systems. Interbull as an EU reference center has a great interest in having this information standardized and therefore comparable. This can lead, via better comparability of information, to harmonization* in methods and definitions between countries.

According to Interbull guidelines, genetic evaluation systems are better to be defined per country per breed per trait group. In current report we collected and compared the information about national evaluation systems, sent to Interbull Centre by the countries participating in the international evaluation for production traits group (milk, fat, protein yields). The current report focuses on the twenty-two (22) countries which are currently EU members, and on dairy production’s traits.

Table 1 Breeds, traits and evaluation models that are included in the dairy cattle genetic national evaluation system of EU members’ countries in addition to United Kingdom, Norway, and Switzerland

bers’ countries in addition to United Kingdom, Norway, and Switzerland

Almost all of the 22 EU countries have a national evaluation for dairy production with the only exception of few of them where a joint evaluation is in place (Table 1). This is the case for: Denmark-Finland Sweden and Germany-Austria-Luxembourg for HOL, JER and RDC breeds; Germany-Austria for BSW breed and Germany-Austria-Czech Republic-Slovak Republic for SIM breed. Country-wise a special case is represented by Belgium where the Walloon Region has a separate evaluation while the Flemish Region is evaluated jointly with Netherlands.

BREED STRUCTURE

As a consequence of its high productivity and Holsteinization-process it is not surprising to see that Holstein (HOL) is the only breed routinely evaluated in all the 22 EU countries (Table 1). Ayrshire together with the other “Red” breeds which have been grouped under the “Red Dairy Cattle” category (RDC) are evaluated in 11 countries: Denmark-Finland-Sweden, Estonia, Germany-Austria-Luxembourg, Ireland, Lithuania, Latvia, and Netherlands-Belgium (Flemish). Simmental breed (SIM) is evaluated in only 10 countries: Croatia, France, Germany-Austria-Czech Republic-Slovak Repiblic, Hungary, Italy, Netherlands-Belgium (Flemish) and Slovenia. Jersey breed (JER) is evaluated in 9 countries: Denmark-Finland-Sweden, Germany-Austria-Luxembourg, Ireland, Italy, and Netherlands-Belgium (Flemish). Brown Swiss breed (BSW) is evaluated in only 4 countries: France, Germany-Austria, Italy, and Slovenia. Within each national evaluation system, a group of breeds can be evaluated either separately i.e. via a single breed evaluation (SB) or jointly, i.e. via a multi-breed evaluation (MB). The latter is applied in Belgium (Walloon), France, Netherlands-Belgium (Flemish), Lithuania and Latvia. Germany does apply a single breed evaluation for SIM and BSW breeds while using a multi-breed evaluation for HOL, RDC and JER breeds. All the remaining 22 EU counties do apply a single breed evaluation (Table 1).

TRAITS

Five traits are generally considered within the dairy production’s trait group, namely Milk, fat and protein yields, and milk fat and protein percent. All the 22 EU countries have evaluation for the three yields traits. Fat and protein percent are additionally evaluated in the following countries: Belgium, Estonia, France, Italy (limited to Jersey), Lithuania, Latvia, and Portugal. All EU countries apply a single-trait evaluation with the except of Belgium "Walloon”, Denmark-Finland-Sweden and Italy (HOL breed only) where such traits are evaluated jointly in a so-called multi-trait evaluation (MT).

DATA RECORDING

All evaluation systems in the 22 EU countries include records from 1st, 2nd and 3rd lactations. Some evaluations systems additionally include records from lactations higher than 3rd , this is the case for Ireland, Netherlands-Belgium”Flemish”, Germany-Austria and Germany-Austria-Czech Republic-Slovak Republic. Looking at the information available, countries do collect such data according to two distinct methods: every 24-hours (method generally known as Test day (TD)) or averaging the production after 305-days of milking (called lactation records).

In the past, often actual TD records were firstly extended into a 305-days lactation records and then included into a particular evaluation model able to account for repeated measures and for this reason known as repeatability (RP) evaluation model. More recently, inclusion of TD records into the evaluation model has been possible using a more complicated model called random regression (RR) model ##(Jamrozik and Schaeffer, 1997)# which allowed inclusion of TD data without the need to extend them into a 305 lactation’s curve. Currently, almost all 22 EU countries make use of the 24-hours records (TD) via a regression model with either random (RR) or fixed (FR) effects***, exception to this, France, Hungary, Italy (Jersey) and Spain where a 305-days records via a repeatability model is still applied.

In conclusion, the most common definition for dairy production traits is milk, protein and fat yields recorded during the first three lactations on a 24-hours interval basis and mostly analyzed applying a regression model, fixed or random.

Figure 1 presents a graphical representation of the most common practices within the 22 EU countries considered by this report.

References

Jamrozik, J., Schaeffer, L.R., 1997. Estimates of Genetic Parameters for a Test Day Model with Random Regressions for Yield Traits of First Lactation Holsteins. Journal of Dairy Science 80, 762–770. https://doi.org/10.3168/jds.S0022-0302(97)75996-4

* standardization is expressing the same, already existing things in the same words, harmonization is changing the procedures/methods/calculations etc. to do the same thing than the others.

*** difference between FR and RR is a difference in data modeling technique. In RR, random effects in addition to some of fixed effects are weighted by weights extracted from curves of lactations across days, whereas in FR, only some of fixed effects are given such weights. Genetic effects are always modeled as random. Almost (but not all) of environmental effects are always modeled as fixed.