Study On Genetic Evaluation Of Chinese Simmental Using External Information

Data structure and method of analysis are two important reasons related to animal genetic evaluation. In terms of progeny testing of cattle, progeny data are usually submitted to different evaluation centers and it is not easy to get all available data of one bull. In such cases, sire rankings differ from one evaluation center to another due to both sampling of progeny data and differences in analytical methods.Therefore, incorporationg external information(EBV and Accuracies) provides another way to improve accuracy other than combining original data. Because external information will be published after each evaluation and everyone can have access to it.Theory of incorporating external information for breeding value estimation was derivated on the bases of Henderson's study in 1975. Computing programme was also compiled using MATLAB language for simulation study. Results from simulation showed that incorporating external information can improve the accuracy of EBVs for common animals. Correlation between EBVs from joint database and incorporative database is 0.99. In another word, incorporative procedure can be used as a substitute for joint procedure.Results of genetic evaluation on Chinese Simmental showed that too limited population size and undesirable data structure are two important problems which affected the accuracy of EBVs on Chinese Simmental. Test run was carried out between Chinese Simmental database and American Simmental database since there have genetic relationship between the two populations. Chinese Simmental was regarded as internal, American Simmental was regarded as external, traits involved in this study were birth weight, weaning weight, yearling weight, milk and maternal weaning weight. Results from joint procedure was taken as standard of evaluation. It is shown that in terms of common animals(22bulls appeared in both populations), incorporative procedure has obvious advantages. Correlation coefficient between EBVs from internal and EBVs from joint is only 0.36, and correlation coefficient between EBVs from incorporative and EBVs from joint increased to 0.92. Whereas, correlation coefficient between EBVs from external and EBVs from joint is 0.99 which is different from simulation. That is because American Simmtal database is quite huge(5,955,372 animals ), Chinese Simmal database only has 3139 animals. Chinese data nearly has no any effect on American Simmental when merging both databases. Incorporative procedure can also improve accuracies of all the other animals in Chinese Simmental population. Correlation coefficient between EBVs from internal and EBVs from joint is 0.528, and correlation coefficient between EBVs from incorporative and EBVs from joint increased to 0.686. Results of this study provide a new method for genetic evaluation in Chinese Simmental.