Back to search

Cornell University

THE APPLICATION OF GENOMIC TECHNOLOGY TO INFER ANCESTRY, IDENTIFY GENETIC RISK FACTORS, AND PREDICT BREEDING VALUES IN WORKING DOGS

Abstract

dc:description.abstract

This dissertation explores how genomic technologies, ranging from direct-to-consumer DNA tests to advanced whole-genome sequencing, can be harnessed to enhance breeding strategies for working dogs. Focusing on two distinct populations, sled dogs and guide dogs, the work demonstrates how a combination of genotype and phenotype data can yield powerful insights into ancestry, disease risk, behavioral traits, and performance outcomes. Beginning with an overview of canine domestication and its implications for specialized working populations, the dissertation then presents case studies on Arctic sled dogs. The work reveals subpopulation structures, admixture patterns, and crucial health trait associations by integrating commercial genotyping data with detailed breed records. Distinct breed ancestries and health traits are identified across the three Arctic sled dog breeds. The research underscores how modern SNP-based methods help identify carriers of genetic disorders and inform selective breeding decisions to preserve crucial genetic diversity. Further chapters pivot to Labrador Retrievers and Labrador–Golden Retriever crosses within guide dog programs, focusing on identifying risk loci on chromosome 2 for primary ciliary dyskinesia. This section illustrates the practical impact of genomic screens and genome-wide association studies (GWAS) in managing heritable diseases, thereby improving working dogs' long-term health and service life. The dissertation then compares the best genomic linear unbiased prediction (GBLUP) and machine learning models (Random Forest, Support Vector Machines, Extreme Gradient Boosting, and Multilayer Perceptrons) to estimate breeding values. Building upon single-trait evaluations, subsequent sections detail the creation of multi-trait selection indices that integrate physical health, behavior, and performance. Across the models, GBLUP showed the highest efficiency and there was no marked increase in performance with denser marker datasets. Indices allowed for multiple traits to be included with single numeric scores and fixed effects using phenotypic data increased model performance. These studies highlight the transformative potential of genomic tools in guiding ethical, evidence-based breeding programs. The findings emphasize that close collaboration among breeders, geneticists, and trainers is paramount to producing healthier, more capable working dogs while safeguarding genetic diversity for future generations.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Animal Science
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Animal Science
Grantor
Cornell University
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Thorsrud, Joseph
Chair dc:contributor.chair
  • Huson, Heather
Committee members dc:contributor.committeemember
  • Boyko, Adam
  • Ivanek Miojevic, Renata
  • Lei, Xingen

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
16938482
Dc Identifier Other
ProQuest Submission ID: 14943
ProQuest Publication ID: 31999717

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Thorsrud, Joseph. THE APPLICATION OF GENOMIC TECHNOLOGY TO INFER ANCESTRY, IDENTIFY GENETIC RISK FACTORS, AND PREDICT BREEDING VALUES IN WORKING DOGS. Doctor of Philosophy thesis, Cornell University, 2025. https://hdl.handle.net/1813/117650