Functional and whole regression-based genome-wide association analyses for weight measurements of chicken eggs

Authors: BURAK KARACAÖREN

Abstract: In genomic studies, complex traits can be modelled using repeated measures, thereby gaining a better understanding of the underlying biology. An increased number of measurements per individual might reduce measurement noise, thus increasing the likelihood of detecting true genomic signals. Here we aimed to predict genomic signals over a logistic curve referring to multiple underlying genetic architectures, for both simulated and longitudinal egg weight datasets. The chicken data were obtained from 92 sires and 801 dams of an 11th generation pure line, resulting in data from 1078 hens. We analysed longitudinal measurements of egg weights with 294,705 single nucleotide polymorphisms (SNPs). We found that a single regression-based functional genome-wide association study (fGWAS) could be useful for manipulating dynamic egg weight over the entire laying period based on a moderate to major effect gene. The fGWAS SNPs associated with the egg weight were located on chromosome 1 close to the gene DLEU7, which has a role in regulating ovary weight in chickens. The SNPs were detected based on the absolute effect sizes using whole regression Bayesian models. This approach is likely to be useful for predicting polygenic risk scores and/or genomic breeding values during the genomic selection/ gene editing for longitudinal egg weight measurements.

Keywords: Egg weight, genomic selection, Gibbs sampling, genome-wide association mapping

Full Text: PDF