Genetic variation between individuals is of great importance for the development of breeding programmes, to select for animals with the most favourable traits. Many production companies routinely measure the feed intake of their animals, in order to calculate efficiency traits such as feed conversion ratio. The development of electronic feeders which automatically record individual intake on a visit-by-visit basis now allows the short-term feeding behaviour of animals to be monitored and analysed as another source of variation between individuals. Due to differences in the resolutions of these feeders as a measurement tool, a standard unit of feeding event needs to be estimated to allow for comparisons between studies. Different models for estimation of the defining value of a meal, the meal criterion, have been used, with the most recent incorporating the change in satiety with time since last feeding as part of the model. In this study I developed a new methodology, based on these models, for use when a within meal population of intervals cannot be easily modelled. I then used this model for application to data from four lines of broiler chickens to estimate meal criteria and compare feeding behaviour within and between the lines. Significant differences were found between fast and slow growing birds, with the faster growing birds having fewer but larger meals than the slower growing birds. However, the lines showed similar structure and bouting of their feeding behaviour, indicating that the fundamental controls of feeding behaviour, such as hunger and satiety, in these lines had been unaltered despite intensive selection for growth. The models were also applied across poultry species, kept in different experimental conditions. A similar structure to the feeding behaviour was found across all these species, with all showing clear separation of feeding events into bouts. In order to estimate the potential use of these behavioural observations in a breeding programme, the heritabilities and genetic correlations with existing performance traits were calculated for the four broiler lines. Heritabilities of all feeding behaviour traits were found to be moderate to high, and very similar across the lines. Correlations with performance traits, however, were low, meaning that there were no clear links of the traits with the current production goals investigated. This indicates that past selection for production has had limited impact on feeding behaviour and also that potential selection for feeding behaviour will have little effect on production gains. To identify the areas of the genome controlling feeding behaviour, traits were associated with a SNP panel. Many regions were found to have highly significant association with feeding behaviour traits, with the most highly correlated traits showing associations with the same regions, suggesting pleiotropic effects of genes in these regions. Future work in this area should include identification of individual genes controlling feeding behaviour to allow prediction of the effects of selection for favourable feeding behaviour on other traits, and comparison of the genotypes of different lines of broilers, to further understand the control of feeding behaviour.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:575891 |
| Date | January 2010 |
| Creators | Howie, Jennifer Ann |
| Contributors | Kyriazakis, Illais; Tolkamp, Berend Jan; Sparks, Nick; Hutchings, Mike; Illius, Andrew |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/7563 |
Page generated in 0.0018 seconds