Beef Cattle Recording: General

Applied beef recording schemes

Beef recording requires recording schemes that can accommodate beef production as implemented in practice. The recording procedures must account for all important effects including the existence of genotype by environment interactions. Beef recording may be undertaken in:

1. Breeding farms.
2. Finishing farms.
3. Individual test stations.
4. Progeny test stations.
5. Abattoirs.

In accordance with existing ICAR terminology recording methods “A”, “B” and “C” may be used to describe the following methods of recording.

1. The A method means recording done by a technician.
2. The B method means recording done by the farmer.
3. The C method means recording done by a mixed system of recording by farmers and technicians.

Factors to be considered

The following factors should be considered as basic requirements in beef recording:

1. A contemporary group may comprise of animals of the same breed, sex and age range kept under the same or at least similar management conditions. Its definition should be carefully established.
2. Tests on animals should be organised in such a way that maximum information can be obtained. This particularly relates to the composition of contemporary groups. This applies especially in relation to the degree of relationship within the contemporary group. The contemporary group animals should be as unrelated as is practically possible.
3. The animals must be identified permanently by a unique number that is always retained with all individual records or documents relating to the animal.
4. Invariant or permanent animal data and further basic information on the animal should be stored in a centralised database. All performance data on an animal should be verified and correct on loading to the database.
5. National cattle databases used to identify, register and monitor birth, movements and the death of animals should be used in the beef recording schemes as far as this is possible.
6. All personnel charged with data collection duties must understand the need for accurate and dated records, which should also include the identification of the recorder. Data may be collected by farm personnel or trained technicians depending on the trait. Complex traits such as conformation assessment using a linear scale or ultrasound measures of fat and muscle must be collected by trained personnel that undergo routine evaluation and retraining procedures when necessary.
7. Data verification systems must be in place which undertake thorough record checks and identify and reject inconsistent or unacceptable data.
8. The contemporary group should include the progeny of at least two sires.

Principles of beef data recording

It is essential that some basic principles should be taken into account in beef recording practices to improve recording efficiency, data storage, data exchange and usability of the animal’s performance data. Throughout the whole recording process, there are four essential key pieces of information which should be included in any animal’s data record:

1. Identification number of the animal.
2. Date of recording.
3. Identification number of the location (farm, station).
4. Identification number of the recorder (recording person).

It is desirable therefore for practical reasons to allocate standardized unique identification codes or numbers not only to the animal but also to locations (holding ID) and to recording personnel. The animal’s holding identification together with recorder identification provides information which allows for the correction for environmental effects and therefore is needed for statistical analysis and genetic evaluation. Furthermore the information in respect of the recorder (recording technician) allows for identification of recording methods (A = recording by official technicians; B = recording by the keeper; C = mixed systems), in accordance with the general ICAR standards.

In general, details relating to an animal can be categorised into four different types as follows.

Invariable data

There are 3 groups of invariable data:

Invariant animal data

This includes all data that are specific to an animal, are available at the birth of the animal and do not change during its lifetime. This set of data comprises at least:

1. The ID no.
2. Birth date.
3. Birth location.
4. Birth type (single, twins, triplets etc.).
5. If the animal is an identical twin or a clone, the ID no(s) of the other genetically identical animal(s).
6. Sex.
7. The breed or breed composition.
8. The ID no. of the animal’s genetic parents.
9. Information in respect of embryo transfer if applicable.
10. ID number of recipient dam in case of embryo transfer.
11. Information in respect of fostering if applicable.
12. ID no. of foster mother case of fostering

Invariant location data

All holdings should have a permanent unique ID to identify correctly fixed effects in genetic evaluation and to study the evolution of these fixed effects (specially herd effects) over time. Furthermore this fixed location ID allows for tracing the origin and later locations of the animal as it moves through the whole production chain.

Invariant recording personnel data

Many records are influenced by an operator or recorder effect. This applies not only for subjective assessments such as linear scoring but also to some degree to measured traits like weights, as the accuracy of the recording and other individual influences differ significantly between recording persons. Therefore, in the case of data recorded by technicians, the operator’s ID number should be included in each record.

Life history data

This class of animal data includes information on the status of the animal (alive or dead, suckling or weaned etc.) and the farm or management conditions the animal is kept in. These data are time-critical in that, for a given animal and a given date, it should be possible to retrieve all relevant information pertaining to management condition, reproduction status etc. There are two main areas of information that have to be collected and permanently updated in this class of data.

Physical location of the animal

Many animals change location during their lifetime. Records may start in the birth herd, continue in a finishing herd or test station and then be completed in an abattoir. The date of arrival and date of departure from each establishment must be recorded so that data collected during each period can be verified if necessary from the recording herd. The identity of an animal must not change between locations. The original identification must be checked before it leaves one location for the next and then checked again on arrival.

The standard format for recording a change of location or status may include the following:

1. Animal ID.
2. Date of change of status/location.
3. Recording person.
4. Current location: farm ID (management-group within farm if applicable).
5. New location: farm ID (management-group within farm if applicable).
6. Range of codes to describe such events as weighed, weaned, died, sold for breeding, sold for slaughter etc.

Animal movements from one herd to another or between management groups within herd, should be recorded as soon as possible.

Reproductive status of the animal

The reproductive status describes the standing of the animal in respect of its breeding cycle/status. It includes such events as mating, insemination, embryo transfer and birth/calving for females, and castration for males. If females are kept with one or several bulls during the mating period, then all possible mates in the mating time window should be recorded. Where natural service is used, then the dates of introduction and withdrawal of sires should be recorded.

The relevant data can also be collected in a standardized format:

1. Animal ID.
2. Date.
3. Recording person.
4. Actual location: farm ID (management-group within farm if applicable).
5. Code to describe the reproductive event.
6. ID of other animal(s) involved (e.g. mating partner, calf, foster calf etc.; if applicable).

Having these two types of data of an animal’s life history, it should be possible to access all relevant information for the calculation of and statistical analysis of performance data.

Recorded data

Recorded data are those details directly recorded on an animal or animal group. It includes both objective measures and subjective assessments.

A number of general principles apply in respect of this data.

1. Provided there is no conflict to legal national units of measurement the data should be recorded in metric units (meters or centimetres, kilogram).
2. All recorded data should be stored as raw data without any adjustment or transformation.
3. Recorded data should include information about all known non-genetic effects and circumstances affecting the level of recorded performance.

It should be noted that a ‘recorded trait’ should strictly be the actual measurement, count or subjective score. If a trait. has to be standardized for a given age or for environmental factors, the resulting adjusted weight is a calculated or derived trait. Adjusted weight may be a function of the recorded weight and age derived from the weighing date and the birth date. Thus, ‘weight’ is a recorded trait, whereas ‘weight at 200 days’ is a calculated or derived trait. In principle, 4 different types of data records can arise.

Objective measurements

Measurements like weights, heights etc. which are assessed with the use of some technical equipment. These measurements, if recorded properly have a high degree of accuracy and are relatively easy to standardize if the definition is clear. However, it should be mentioned that some recording device (e.g. ultrasound measurements) needs careful training and supervision of the operator as otherwise the accuracy of measurement is not guaranteed.

Date/Time

It is strongly recommended that for recording purposes, the recording date should be used rather than the animal’s age. The reason is, that additional information is required to derive the age of an animal, and this may lead to erroneous recordings, arising from different formats (age in years, months, or days) or just deficient or inaccurate information which can subsequently be corrected. The recording date allows for the calculation of age when combined with the birth date. The birth date should be recorded in the database for every animal.

The date of recording also provides information on the month or season in which the recording has been undertaken in. This information may be useful for the further interpretation or statistical/genetic analyses of the recorded data. Where date of data collection is recorded then the date should be stored as an 8-digit number using the format

 YYYYMMDD.

For most performance traits the date of data collection is sufficient information, the time not normally being necessary unless needed for management reasons. However, where recording time is collected then the 24-hour clock should be used. The time should be stored as an 6-digit number using the format

 hhmmss.

Nominal classification

This occurs where observations are recorded in discrete, unordered classes, like breed or reason for disposal. Well defined and comprehensive categories are required to gain as much information as possible. The classes should be mutually exclusive, i.e. no overlapping of classes should occur. There may be a need for an additional open class for all cases, that cannot be attributed to one of the defined classes. This class should be as small as possible and should include a brief description in order to facilitate the creation of additional classes if necessary.

Subjective scores

This type of recording classifies animals, using a finite ordinal scale, into one of a number of possible classes. Often the classes are an ordered sequence of numeric scores, where the lowest and the highest numbers represent extreme phenotypes expressed in the population under consideration.

It is desirable that descriptions of the different classes be provided in text and where appropriate as pictures/drawings. As outlined later, the main problem with subjective scores is to ensure that values are comparable, even if they are assessed by different persons or by the same person at different points of time and at different locations. This requires clear definitions, ongoing and systematic training as well as the permanent supervision of the recording process. It is essential that periodic verification of the aptitude of the recording technicians be undertaken.

Regardless of the type of recorded trait, it is possible to use a standard format:

1. Animal ID (or group of animals if applicable).
2. Date of recording.
3. Recording person.
4. Actual location: farm ID (management-group within farm if applicable).
5. Trait name/trait code.
6. Trait value.
7. Additional information pertaining to the animal.
8. Additional information pertaining to the recording procedure.

It is essential, that for all recorded traits in a given recording scheme, the trait be sufficiently well defined. Additionally unique two or three letter trait codes may be specified (e.g. one code for “shoulder width”, another code for “roundness of thighs” etc.) where it is not practical to use the full name. It is strongly recommended to use trait definitions and/or trait codes in accordance with international standards where available from an international breed umbrella organisation.

Calculated traits

This type of trait is different from the other categories, as calculated traits are derived from the ‘raw’ data information. These traits are calculated according to clearly defined rules. Where the calculated trait requires complex computing procedures or is frequently used, the results may be stored rather than re-calculated each time.

In general, calculated traits may be divided into three different classes of traits.

Counts

This category include summarized information from recordings such as the number of inseminations or matings per mating period, the number of calves born and the number of ticks observed per unit area.

Adjusted or derived traits

Raw data will often have to be adjusted to a defined age, weight, or length of testing period, to comply with the defined standard. If, for example, the weight at 365 days is defined as a standard beef trait, but an animal which is born on March 1, 2000 is weighed on March 15, 2001, the recorded weight is taken at 380 days. Therefore, it has to be adjusted to the standard age by using a linear or other adjustment procedure.

For these classes of traits it makes sense to use a similar data format as for the unadjusted recorded trait. Distinct trait codes should be used in order to avoid confusion. Information that already has been accounted for in any adjusting procedure is omitted.

Functions of several recorded traits

A number of interesting performance traits are derived from a combination of recorded traits. Daily gain in the test period for example is the difference between weight at end and weight at start of the test period, divided by the difference of age at end and age at start of test period, expressed as grams per day. This type of data can be derived both from raw recorded data and from adjusted traits. With these kind of traits, one often has several overlapping additional pieces of information. For example combined traits are recorded by different recorders, at different dates, and at different locations. Combined traits therefore should be defined to be largely independent of this type of additional information. A daily gain in a test period should pertain to a standardized test length. The trait definitions given in the following section will specify which additional information is needed in detail. The resulting general data format for calculated traits may be as follows:

1. Animal ID (or group of animals if applicable).
2. Date of recording (start/end of test period etc.).
3. Age of animal.
4. Relevant location.
5. Trait code of calculated trait – where applicable.
6. Value of calculated trait.
7. Additional information pertaining to the animal (e.g. contemporary group).

Note that in this case the age (as a calculated trait) is included, while for recording purposes, it is strongly recommended to record actual dates of events.

Genetic proofs and other population-related indices

This type of data applies, if an animal’s performance is related to the performance of other animals in the same population. Genetic evaluation includes trait information (raw or adjusted), pedigree information, classification of fixed environmental effects and covariables etc. Typically such analyses are done for all animals of a population simultaneously.

Results of genetic proofs are by definition independent of any environmental factors, but values may change over time. Therefore they should be stored with the animal’s identification number, the date of estimation together with a definition of the reference base used in the particular genetic evaluation.

Data requirements for the calculation of genetic proofs

In most cases the required data formats for trait information, fixed and random effects and pedigree information are clearly defined in the genetic evaluation system. The data file should be provided in a standard format. Where raw data is subjected to ongoing maintenance which allows for changes of historical data (e.g. change of parentage, fixed effects etc.), submitted data for genetic evaluation should include all animals of the relevant population rather than just a subset of new or recently recorded animals. Data for the calculation of genetic proofs should comprehensively account for management conditions and other non-genetic effects affecting the animal’s performance. Much attention should be paid in the definition of contemporary groups held under similar management conditions. However, the definition of contemporary groups frequently will be a compromise between a precise specification of the group with possibly loss of contemporaries on the one hand and a wider specification with loss of information accounting for fixed effects. Usually the pedigree file is a separate file containing the animal identification number and that of its parents together with breed sex and birth date. The pedigree file should contain all animals contributing to the genetic structure of the breeding population. Where pedigree data originates from separate regional or historical sub-populations or separate databases, it may happen that different ID numbers and/or different names of identical animals occur. Therefore special consideration should be given to identifying and attaching unique ID numbers to the relevant animals.

There are some special situations which need to be taken into account:

1. In case of identical twinning and cloning, it is necessary to record the fact that two or more individuals are genetically identical, since on the basis of pedigree information alone (identical parent IDs), these animals would be falsely identified as full sibs.
2. In genetic evaluation systems it is common practice to include ‘genetic groups’ for founder animals. Animals with unknown parents are grouped according to age (year born), country of origin and/or breed composition (if more than one breed is included). Therefore, it is essential to record this data especially for older animals in the pedigree file.

Data storage and management

Given that genetic proofs will be used for the assessment of the production or breeding potential of an animal, it is essential that data are stored in a centralized form, which typically would be a national database, but also may be a database at the level of regions, large farms, commercial breeding companies or breed associations etc. The necessity for a database results from the fact, that performance data of different animals or the same animal at different ages might be combined to retrieve the relevant information. Ideally, data from one ‘breeding population’ are stored in one database or in databases following a common structure with well established links and defined interfaces for data exchange. The data structure should be defined in such a way, that flexible and efficient use of the relevant data for a variety of purposes is enabled. ‘Structure’ means both the hierarchy of different types of data and the general format, in which data should be recorded and stored.