Register      Login
Animal Production Science Animal Production Science Society
Food, fibre and pharmaceuticals from animals
Shopping Cart: ( empty )
You are here: Home > Journals > AN > EA07356 > Fulltext
RESEARCH ARTICLE
Contents Vol 48(8)

Designing complex research projects to estimate genetic parameters plus treatment and other effects – optimising the experimental design

D. L. Robinson
+ Author Affiliations
- Author Affiliations

Primary Industries Innovation Centre, NSW Department of Primary Industries, University of New England, Armidale, NSW 2351, Australia. Email: dorothy.robinson@dpi.nsw.gov.au

Australian Journal of Experimental Agriculture 48(8) 1110-1117 https://doi.org/10.1071/EA07356
Submitted: 1 October 2007  Accepted: 21 December 2007   Published: 14 July 2008

Abstract

There is an increasing trend towards integrated research, in which several individuals or institutions pool their expertise and make use of common resources, collaborating towards a common set of scientific goals. Integrated research enables a larger number of factors to be investigated, and the most influential or important ones identified, providing information on how the different factors interact or fit together. Good experimental design is, however, required to ensure the aims can be achieved and resources spent wisely.

Issues involved in the experimental design of the Australian Beef Cattle Cooperative Research Centre for Meat Quality are discussed. Theoretical results and simulation studies were used to determine optimal numbers of progeny per sire for estimating genetic parameters. For heritabilities of 0.2 and 0.5, the optima are respectively 21 and 9 progeny with recorded measurements. The curves surrounding the optima are quite flat, so aiming for 10–15 progeny with measurements per trait should provide reasonable accuracy in many situations. Estimates of heritabilities, genetic correlations and phenotypic variances have lower sampling correlations than genetic variances and covariances, suggesting that when results are pooled over different breeds or trials, it is better to pool estimates of heritabilities and genetic correlations than (co)variances.

Using sires in more than one year increases the robustness of estimated sire effects and increases the accuracy of genetic parameter estimates for hard-to-measure traits (e.g. feed efficiency) that are not recorded on all animals. Unless sires can be chosen as a true random sample of the population, arrangements of link sires (and other effects such as treatments) should be chosen to provide accurate estimates when all terms in the model are fitted as fixed.


References


Foulley JL , Bouix J , Goffinet B , Eisen JM (1990) Connectedness in genetic evaluation. In ‘Advances in statistical methods for genetic improvement of livestock’. (Eds D Gianola, K Hammond) pp. 277–308. (Springer-Verlag: New York)

Gilmour AR , Gogel BJ , Cullis BR , Welham SJ , Thompson R (2002) ‘ASReml user guide. Release 1.0.’ (VSN International Ltd: Hemel Hempstead, UK)

Henderson CR (1984) ‘Applications of linear models in animal breeding.’ (University of Guelph: Guelph, Canada)

Johnston DJ, Tier B, Graser H-U, Girard C (1999) Presenting BREEDPLAN version 4.1. Proceedings of the Association for the Advancement of Animal Breeding Genetics 13, 193–196. open url image1

Meyer K (1991) Estimating variances and covariances for multivariate animal models by restricted maximum likelihood. Genetics, Selection, Evolution. 23, 67–83.
| Crossref | GoogleScholarGoogle Scholar | open url image1

Press WH , Teukolsky SA , Vetterling WT , Flannery BP (1992) ‘Numerical recipes in Fortran 77.’ 2nd edn (Cambridge University Press: Cambridge)

Robertson A (1959a) Experimental designs in the evaluation of genetic parameters. Biometrics 15, 219–226.
| Crossref | GoogleScholarGoogle Scholar | open url image1

Robertson A (1959b) The sampling variation of the genetic correlation coefficient. Biometrics 15, 469–485.
| Crossref | GoogleScholarGoogle Scholar | open url image1

Robinson DL (1987) Estimation and use of variance components. The Statistician 36, 3–14.
| Crossref | GoogleScholarGoogle Scholar | open url image1

Robinson DL (2002) Applications of statistical design and analysis to genetics. PhD Thesis, University of New England, Armidale.

Robinson DL (2008) Experimental design for complex, large-scale research projects to estimate genetic parameters plus numerous treatment and sire effects. Livestock Science, in press.

Upton W, Burrow HM, Dundon A, Robinson DL, Farrell EB (2001) CRC breeding program design, measurements and database: methods that underpin CRC research results. Australian Journal of Experimental Agriculture 41, 943–952.
| Crossref | GoogleScholarGoogle Scholar | open url image1