BORIS Theses

BORIS Theses
Bern Open Repository and Information System

Genetic analysis of female fertility focussing on multiple birth events in Swiss cattle

Widmer, Sarah (2022). Genetic analysis of female fertility focussing on multiple birth events in Swiss cattle. (Thesis). Universität Bern, Bern

22widmer_s.pdf - Thesis
Available under License Creative Commons: Attribution-Noncommercial-No Derivative Works (CC-BY-NC-ND 4.0).

Download (16MB) | Preview


In dairy cattle farming, intensive selection for milk yield has led to a decline in female fertility in the last decades, due to unfavourable genetic correlations between milk yield and female fertility. Phenotypes included in current genetic evaluations of fertility are interval and binary traits, calculated from insemination and previous calving date records and deduced from insemination success, respectively. For improved selection, the development of novel phenotypes that describe the physiology of reproduction more precisely would be beneficial. A potential novel phenotype is multiple births. Especially in dairy cattle, multiple birth events are undesirable due to negative impacts on a cow’s performance and potential health issues of the dam and the calves. In the first part of the thesis, I investigated the genetic background of multiple birth events in population studies for the four main Swiss dairy cattle breeds. For this purpose, I designed a breeding value estimation for this novel phenotype in Switzerland. By applying genome-wide association studies (GWAS) on the estimated breeding values, quantitative trait loci (QTL) for multiple births were detected in the three different Swiss dairy cattle populations Holstein, Brown Swiss and Original Braunvieh on chromosomes 11, 15 and 11, respectively. In all populations I identified candidate causal variants affecting the expression of the genes LHCGR, FSHR, ID2, PRDM11 and SYT13 by using linkage disequilibrium analysis for fine-mapping. In the second part of the thesis, I tested alternative methods to identify associated genomic regions, which do not require a complex pipeline of specialised software-tools and massive computing resources. Preliminary work for using machine learning tools in the analysis of binary traits was provided. Thereby, I used the Least Absolute Shrinkage and Selection Operator (Lasso), support vector machine and random forest algorithms for identifying QTL in a case/control approach. The machine learning approaches were validated as promising and efficient alternatives to classical methods. Their application led to the identification of genomic regions showing suggestive associations for multiple births in Holstein cattle. In future, the machine learning tools random forest, Lasso and support vector machine can offer a low input alternative for GWASs while the availability of data for traits of interest are increasing. The identification of QTL for multiple births improves the understanding of the genetic architecture that underlies our trait of interest and female fertility in general. By developing the breeding value prediction, we set the foundation for implementing our knowledge in the breeding strategies to avoid multiple births in future. Considering this novel phenotype of female fertility will improve the sustainability of dairy cattle farming.

Item Type: Thesis
Dissertation Type: Cumulative
Date of Defense: 21 December 2022
Subjects: 500 Science > 590 Animals (Zoology)
600 Technology > 610 Medicine & health
600 Technology > 630 Agriculture
Institute / Center: 05 Veterinary Medicine > Department of Clinical Research and Veterinary Public Health (DCR-VPH) > Institute of Genetics
Depositing User: Hammer Igor
Date Deposited: 27 Apr 2023 13:06
Last Modified: 21 Dec 2023 23:25

Actions (login required)

View Item View Item