Full text in pdf format

Practical Guidelines for Analyzing Data Under an Animal Model with DFREML Using Sparse Matrix Solver. L Univariate Analysis

J.B.S. Ferraz*

Departamento de Produção Animal, Faculdade de Medicina Veterinária e Zootecnia, USP, Caixa Postal 23, 13630 Pirassununga, SP, Brasil
* Present address: University of Nebraska, Dept. of Animal Science, Lincoln, NE 68583-0908, USA

ABSTRACT

The best estimation of variance components and genetic parameters in the unbalanced data usually used in anima1 breeding is that obtained by restricted maximum likelihood (REML) procedures under an anima1 model that consider all the pedigree information available; a model that is as close as possible to the biologica1 situation. The use of a derivative-free algorithm, where the inversion of the big matrices generated in anima1 models is not needed and the use of a sparse matrix solver - SPARSPAK- resulted in a software that solves the complex mixed model equations in a shorter time and using less computer resources than the origina1 version of DFREML. This paper presents the practical guidelines for animal breeders that need to use this package but are not familiar with FORTRAN or the previous version of DFREML.

Keywords: Guidelines; Univariate analysis.

REFERENCES

George, A., Liu, J. and Ng, E. (1980). User guide for SPARSPAK: Waterloo sparse linear equations package. CS-78-30, Depto. Computer Science, Univ. of Waterloo, Ontario, Canada.

Graser, H.U., Smith, S.P. and Tier, B. (1987). A derivative-free approach for estimating variance components in animal models by Restricted Maximum Likelihood. J. Anim. Sci. 64: 1362-1370.

Henderson, C.R. (1949). Estimation of changes in herd environment. J. Dairy Sci. 32: 706.

Henderson, C.R. (1988). Theoretical basis and computacional methods for a number of different anima1 models. J. Dairy Sci. 71 (Suppl.): 1-16.

Henderson, C.R., Kempthome, O., Searle, S.R. and Von Krosigk, C.M. (1959). The estimation of genetic and evironmental trends from records subject to culling. Biometrics 15: 192-218.

Kennedy, B.W., Schaeffer, L.R. and Sorensen, D.A. (1988). Genetic properties of animal models. J. Dairy Sci. 71 (Suppl.): 17-26.

Meyer, K. (1988a). DFREML - A set of programs to estimate variance components under an individual Animal Model. J. Dairy Sci. 71 (Suppl.): 33-34.

Meyer, K. (1988b). DFREML - Programs to Estimate Variance Components for Individual Animal Models by Restricted Maximum Likelihood (REML).- USER NOTES. University of Edinburgh.

Meyer, K. (1989). Estimation of variance components for individua1 Animal Models. I. Univariate analyses. Genet. Sel. Evol 21: 317-340.

Meyer, K. (1991). Estimation of variance components for individual Animal Models. II. Multivariate analyses. Genet. Sel. EvoL 23: 67- 83.

Quaas, R.L. and Pollak, E.J. (1980). Mixed model methodology for farm and ranch beef cattle testing programs. J. Anim. Sci. 51: 1277-1287.

Van Vleck, L.D. and Nunez-Dominguez, R. (1990). Genetic evaluation of Dairy herds bulls and cows with an animal model. In: International Dairy Seminar. Proceedings. Centro de Ganaderia, Colegio de Postgarduados, Chapingo, Mexico.