Degree Type

Dissertation

Date of Award

2019

Degree Name

Doctor of Philosophy

Department

Animal Science

Major

Animal Breeding and Genetics

First Advisor

Jack C. Dekkers

Abstract

Resilience is defined as the ability of an animal to maintain a high level of performance in any environment. Swine breeding companies are organized in a pyramid-like structure. As a result of management differences between the different tiers of the pyramid, genotype-by-environment (GxE) interactions have become a much bigger issue that continues today. Genetic selection for resilience can be accomplished in at least two ways, i) measuring indicator traits in the nucleus, such as antibody levels, and ii) measuring performance phenotypes in commercial testing herds. Both of these techniques provide opportunities to improve resilience in livestock.

Total lost productivity in swine in an ideal versus heavily challenged commercial environment has been estimated at 30%, indicating a large opportunity to close this gap. Porcine reproductive and respiratory syndrome (PRRS) costs the swine industry an estimated $664 million annually. Forty-five percent of this cost is due to economic losses in breeding herds. The objective of this thesis was to quantify resilience in sows using an indicator trait, PRRS antibody levels in blood after an outbreak, and to quantify and determine the genetic basis of resilience in growing pigs using individual daily feed intake data.

Antibody levels were quantified using blood from sows during a PRRS outbreak in a nucleus herd, as an indicator trait to improve resilience. Antibody level was estimated to have a heritability of 0.17±0.05 and low genetic correlations with litter size traits during the outbreak, which conflicts a previous study, except for the genetic correlation with the number of stillborns, which was -0.73±0.29. Future research is needed to understand why antibody level in sows after a PRRS outbreak was not a good genetic indicator of resilience in this study.

Collecting commercial data directly for selection is a viable alternative to using indicator traits to quantify resilience. The current thesis quantified resilience phenotypes from individual daily feed intake data in a natural disease challenged environment. The root mean square error (RMSE) phenotype quantified within animal variation in feed intake over time. The quantile regression (QR) phenotype quantified the proportion of off-feed days. Finally, the run of depression (ROD) phenotype quantified the proportion of days that fall in consecutive stretches below the expected feed intake. These novel resilience phenotypes can be used to quantify and select for resilience, which was validated with heritability and genetic correlation estimates with mortality and treatment rate. Further modifications of these phenotypes were tested, but many of these failed to improve the phenotype in terms of heritability and genetic correlations with mortality and treatment rate. As the use of precision agriculture becomes more common in the commercial livestock industry, genetic evaluation systems will need to find ways to harness these data to improve resilience and other traits.

Copyright Owner

Austin Michael Putz

Language

en

File Format

application/pdf

File Size

271 pages

Share

COinS