Use this URL to cite or link to this record in EThOS:
Title: The design of structured pig breeding programmes
Author: Grundy, Brian
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 1993
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The aim of this thesis was to investigate the features that underpin the Group Nucleus breeding scheme for pigs in which the population is subdivided into several herds. With Best Linear Unbiased Prediction it is possible to directly assess response. The estimation is, however, dependent on the underlying variance components used. In general, estimating breeding values with an inflated heritability in the model results in a high predicted response, whilst having much less of an effect on the actual response. Additionally, the reduction in the weight of family information results in more unrelated animals being selected. A method to utilise this effect in order to reduce inbreeding is presented. As the population is subdivided across farms, analyses were undertaken to determine genetic and phenotypic parameters both within and across farms; little heterogeneity of variance occurred for litter size. The low heritability of the trait does however confirm the need for specialised selection methods in order to achieve satisfactory response. The production traits also showed low heritabilities, but with up to twofold differences between farms. Further analyses of the data indicated that this heterogeneity of variance was due in part both the environmental differences and a sire by farm environment interaction. The effect of altering the proportion of artificial insemination (AI) to link farms was investigated. In general, the rate of response is robust to changes in proportion of AI matings for all but the lowest proportion AI, mainly because both AI boars and natural service boars (only used in a single herd) are highly selected. Moveover, the increased number of boars associated with natural service, for example at 90% compared to 100% AI, can yield greater responses in the long term due to a lower rate of inbreeding and consequently a larger available additive genetic variance. In summary, the theoretical studies indicate that a Group Nucleus population is a robust system in which to implement genetic selection with alternative testing procedures, proportion AI or parameter use effective for all but the most extreme cases. In practice, however, additional factors can cause low heritability estimates and subsequently low rates of predicted responses, and these are discussed.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available