AI bull evaluation standards for dairy and dual purpose breeds

Abstract

The purpose of this report is to recommend a code of practice to be followed by AI organisations and sire evaluation centres. The genetic merit of AI sires should be expressed as Breeding Values (BV) or as Relative Breeding Values (RBV). This value should be calculated for all traits which are considered for sire selection. The methods of estimation of the BV are discussed and it is suggested that where possible, methods based on the solution of properly constructed simultaneous equations (e.g. BLUP) should be used. For these new methods, a clearly defined reference base is required. It is recommended that where possible, this should be the average BV of the latest batch of bulls completing progeny test. If a fixed base is used, the average merit of bulls completing tests each year should be available. Besides good evaluation procedures and correction factors it is most important to ensure random use of the young bulls in many herds when they are doing their test inseminations in order to get unbiased progeny test figures. Only reliable figures should be published: the minimum requirements are suggested. A list of economically important traits is discussed. The traits are grouped under three headings: production traits, reproduction traits and management traits. Possible systematic environmental factors which have to be considered when estimating Breeding Values are listed in Appendix 1. The calculation of a selection index is recommended for use as a guide for proven bulls. The index can only be used in areas where the applied economic factors are appropriate, and so it is important to publish BV's for each trait separately. It is important that all test results should also be published in official bulletins. It is suggested, however, that the BV of important traits should not be published if the repeatability ( r 2) is less than 65%; for less important traits the repeatability of the estimated BV should be at least 35%. A scheme is suggested as to how the proofs can be presented. For international semen exchange the Relative Breeding Value is the recommended unit of evaluation. This value has the advantages that it is fairly independent of the mean levels and unit of measurement, and it is similar for all traits. A list of additional information which should be available is provided. Some important items are mentioned that should be taken into consideration when trials are planned, to compare different populations.