Mutant alleles differentially shape cattle complex traits and fitness

Abstract

Classical mutations tend to be deleterious to traits and fitness. Is this the case for mutations with polygenic effects? Here, we infer ancestral and mutant alleles (MAs) for 8 million sequence variants in 113k cattle and quantify the effects of MA on 37 complex traits. Heterozygosity at sites conserved across 100 vertebrates increase fertility, stature, and milk production, positively associating these traits with fitness. MAs decrease fat and protein concentration in milk and stature but increase gestation length and somatic cell count in milk indicative of mastitis. However, the frequency of MAs that decrease fat and protein concentration and stature and increase gestation length and somatic cell count is lower than the frequency of MAs with the opposite effect. These results suggest bias in the direction of effect of mutation (e.g. towards reduced protein in milk), but selection operating to reduce the frequency of these MAs. MAs with a large-effect decrease protein and milk yield, while small-effect MAs increase the two traits. These results imply two classes of genomic sites subject to long-term selection: sites conserved across vertebrates show hybrid vigour while sites subject to less long-term selection show a bias in mutation towards alleles that are selected against.

Gene expression and RNA splicing explain large proportions of the heritability for complex traits in cattle

“…The details of the imputation were described previously 38 . Briefly, the imputation of biallelic sequence variants was performed with Minimac3 39,40 and those variants with imputation accuracy R 2 > 0.4 and minor allele frequency (MAF) > 0.005 in both bulls and cows were kept.…”

“…Phenotype data included a total of 8,949 bulls and 103,350 cows, including Holstein (6,886♂ / 87,003♀), Jersey (1562♂ / 13,353♀), cross-breed (36♂ / 5,037♀) and Australian Red (265♂ / 3,379♀) dairy breeds. In total, 37 traits were studied that related to milk production, mastitis, fertility, temperament and body conformation and the details of these traits can be found in 38 . For AVR blood samples breed and days in milk (DIM) were fitted as fixed effects in the model.…”

“…The genotype data for Australian animals including those used for e/sQTL mapping (blood and milk cells) and association analysis of phenotypes (described later) were 16,251,453 sequence variants imputed using Run7 of the 1000 Bull Genomes Project 36,37 . The details of the imputation were described previously 38 . Briefly, the imputation of biallelic sequence variants was performed with Minimac3 39,40 and those variants with imputation accuracy R 2 > 0.4 and minor allele frequency (MAF) > 0.005 in both bulls and cows were kept.…”

Genetic score omics regression and multi-trait meta-analysis detect widespread cis-regulatory effects shaping bovine complex traits

“…Project 47,48 . The details of the imputation were described previously 49 . Briefly, the imputation of bi-allelic sequence variants was performed with Minimac3 including Holstein (6,886♂ / 87,003♀), Jersey (1562♂ / 13,353♀), cross-breed (36♂ / 5,037♀) and Australian Red (265♂ / 3,379♀) dairy breeds.…”

“…Briefly, the imputation of bi-allelic sequence variants was performed with Minimac3 including Holstein (6,886♂ / 87,003♀), Jersey (1562♂ / 13,353♀), cross-breed (36♂ / 5,037♀) and Australian Red (265♂ / 3,379♀) dairy breeds. In total, 37 traits were studied that related to milk production, mastitis, fertility, temperament, and body conformation and the details of these traits can be found in 49 . For AVR blood samples, breed and days in milk (DIM) were fitted as fixed effects in the gene expression and splicing GWAS model.…”