Using mapped quantitative trait loci in improving genetic evaluation

Date
2001-01-01
Authors
Abdel Azim, Gamal
Major Professor
Advisor
Albert E. Freeman
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Animal Science
Organizational Unit
Journal Issue
Series
Department
Animal Science
Abstract

The benefit of using QTL information in dairy cattle breeding schemes by means of computer simulation is investigated. In addition, algorithms to overcome computational problems arising when marker data are included in mixed linear models were proposed.;Computer simulation was conducted with parameters relative to the Holstein population of the United States. Superiority of QTL-assisted selection (QAS) over QTL-free selection was studied in four pathways of selection, namely active sires, young bulls, bull dams, and cows, for cumulative genetic response, accuracy of evaluation, and selection pressure on the QTL.;Further, breeding scheme as a factor was studied. The breeding scheme was the most effective factor in increasing the superiority of QAS. As it agreed with many previous studies, nucleus breeding schemes were found to be promising systems to implement QTL information. On the other hand, benefits of QAS in conventional two stage selection programs were limited.;The interaction between the type of QTL information available and the breeding system was found important. Using a highly polymorphic QTL in nucleus schemes was found very effective. Effects of different number of alleles per locus and different number of loci on the superiority of QAS were studied.;An algorithm to directly build the inverse of a conditional gametic relationship matrix, given marker data, was developed. The inverse algorithm is based on matrix decomposition instead of partitioned matrix theory. Numerical techniques that greatly improved computing performance were introduced.;Appropriate modifications to the conventional breeding schemes that are currently in use are highly recommended. Further, attention should be paid to the characteristics of the QTL and how they may interact with the breeding system, e.g., number of loci and alleles. Finally, the study found that the use of marked or known QTL information in genetic evaluation is computationally possible and generally useful.

Comments
Description
Keywords
Citation
Source