Untersuchung zur Gestaltung von Zuchtprogrammen in der Legehennenzucht

Abstract

The primary objective of the present study was to analyse the conventional layer breeding. Among others, inbreeding, selection response, different selection and mating strategies were evaluated for this purpose. Finally, modern methods and optimization concepts were designed to improve the layer breeding programs based on the research results of conventional layer breeding schemes. Three lines of commercial Lohmann LSL layers were used for the evaluation. For these lines, different scenarios were simulated, including the mating system, the number of selected hens and cocks as well as homozygosity. Selected number of hens and cocks was modified to test the effect on selection response.Three groups of 500, 1500 and 4,000 animals in single cage housing systems were used for the simulation study. The study showed that in a large flock of 4000 single cages, highest selection response was achieved by selecting 500 hens and 50 cocks. In a medium flock of 1500 single cages, the selection of 200 hens and 50 cocks were optimal. In a small (reserve) flock of 500 single cages, high selection response could be achieved by selecting 120 hens and 30 cocks.The comparison of different mating systems showed that the average inbreeding coefficient when using assortative mating was 91% higher compared to that of random mating. However, assortative mating resulted in an increase in genetic response of only 5% compared with random mating. The genetic response and inbreeding with minimum coancestry mating and disassortative mating were below random mating. Therefore, regarding selection response and inbreeding, the choice of the appropriate mating system is of great importance.The simulation showed, based on the population under study, the loss of alleles within 10 generations. After 10 generations of selection, from originally 100 alleles of 50 founder sires and 1000 alleles of 500 founder dams, respectively, remained less than only18 different alleles in the large flock.In this study, the optimum genetic contribution (OGC) theory (Meuwissen, 1997) was applied to maximize the genetic response at a predefined rate of inbreeding. The theory is based on equalising the genetic contribution of the founder animals to the following generation. This could be implemented by the software package GENCONT. Under certain constraints, the optimal number of offspring for each candidate can be determined using this programme, to restrict the inbreeding rate and to balance the different contribution. Generally, the restriction can be carried out on the basis of average relationship or average inbreeding. As an alternate scenario, the restriction was carried out through predefining the number of offspring per hen and cock, respectively. Finally, the three scenarios were compared with the standard method of the commercial breeding company. The result showed that high rate of genetic gain with limited inbreeding could be achieved when using OGC. The alternate scenario resulted in decreased breeding value for all the evaluated lines. There was no difference in genetic response whether the maximum tolerated relationship was predefined on the basis of average relationship or average inbreeding. Taking the overlapping generation into account turned to have no remarkable effect.The potential of marker assisted selection was evaluated using simulation studies. Three types of selection strategies were compared: (1) phenotypic selection (PAS): entirely based on phenotypic information (genotype information was not considered); (2) gene assisted selection (GAS): selection using information on the QTL; and (3) marker assisted selection (MAS): selection using information on markers linked to the QTL. The comparison of PAS and GAS showed only a short-term (i.e. less than 5 generation) advantage of GAS over PHE. GAS resulted in an increase of genetic gain by up to 20,6%. However the superiority of GAS declined in the long-term (i.e. greater than 5 generation). The genetic gain achieved with MAS and PAS was identical.The effect of number of alleles per marker and length of chromosome segment was also evaluated. Simulations were carried out for three types of selection strategies (PAS, MAS and GAS) during 10 generations. At first, the number of alleles per marker was increased from 2 to 10. However, extra genetic gain couldn t be achieved when increasing the number of marker alleles. Increasing the length of the chromosome segment didn t have any effect on genetic gain either. This suggests that the information content of each marker genotype is more important than the number of alleles per marker and length of chromosome segment, respectively.