Abstract
Genome-wide association studies are routinely used to identify genomic regions associated with traits of interest. However, this ignores an important class of genomic associations, that of epistatic interactions. A genome-wide interaction analysis between single nucleotide polymorphisms (SNPs) using highly dense markers can detect epistatic interactions, but is a difficult task due to multiple testing and computational demand. However, It is important for revealing complex trait heredity. This study considers analytical methods that detect statistical interactions between pairs of loci. We investigated a three-stage modelling procedure: (i) a model without the SNP to estimate the variance components; (ii) a model with the SNP using variance component estimates from (i), thus avoiding iteration; and (iii) using the significant SNPs from (ii) for genome-wide epistasis analysis. We fitted these three-stage models to field data for growth and ultrasound measures for subcutaneous fat thickness in Brahman cattle. The study demonstrated the usefulness of modelling epistasis in the analysis of complex traits as it revealed extra sources of genetic variation and identified potential candidate genes affecting the concentration of insulin-like growth factor-1 and ultrasound scan measure of fat depth traits. Information about epistasis can add to our understanding of the complex genetic networks that form the fundamental basis of biological systems.
| Original language | English |
|---|---|
| Journal | Journal of Animal Breeding and Genetics |
| Volume | 132 |
| Issue number | 2 |
| Pages (from-to) | 187-197 |
| Number of pages | 11 |
| ISSN | 0931-2668 |
| DOIs | |
| Publication status | Published - 1 Apr 2015 |
Keywords
- Faculty of Health and Medical Sciences
- Genetics
- quantitative genetics
- Faculty of Science
- Genetics
- quantitative genetics