Overlap in genomic variation associated with milk fat composition in Holstein Friesian and Dutch native dual-purpose breeds

Abstract

The aim of this study was to identify if genomic variations associated with fatty acid (FA) composition are similar between the Holstein-Friesian (HF) and native dual-purpose breeds used in the Dutch dairy industry. Phenotypic and genotypic information were available for the breeds Meuse-Rhine-Yssel (MRY), Dutch Friesian (DF), Groningen White Headed (GWH), and HF. First, the reliability of genomic breeding values of the native Dutch dual-purpose cattle breeds MRY, DF, and GWH was evaluated using single nucleotide polymorphism (SNP) effects estimated in HF, including all SNP or subsets with stronger associations in HF. Second, the genomic variation of the regions associated with FA composition in HF (regions on Bos taurus autosome 5, 14, and 26), were studied in the different breeds. Finally, similarities in genotype and allele frequencies between MRY, DF, GWH, and HF breeds were assessed for specific regions associated with FA composition. On average across the traits, the highest reliabilities of genomic prediction were estimated for GWH (0.158) and DF (0.116) when the 8 to 22 SNP with the strongest association in HF were included. With the same set of SNP, GEBV for MRY were the least reliable (0.022). This indicates that on average only 2 (MRY) to 16% (GWH) of the genomic variation in HF is shared with the native Dutch dual-purpose breeds. The comparison of predicted variances of different regions associated with milk and milk fat composition showed that breeds clearly differed in genomic variation within these regions. Finally, the correlations of allele frequencies between breeds across the 8 to 22 SNP with the strongest association in HF were around 0.8 between the Dutch native dual-purpose breeds, whereas the correlations between the native breeds and HF were clearly lower and around 0.5. There was no consistent relationship between the reliabilities of genomic prediction for a specific breed and the correlation between the allele frequencies of this breed and HF. In conclusion, most of the genomic variation associated with FA composition in the Dutch dual-purpose breeds appears to be breed-specific. Furthermore, the minor allele frequencies of genes having an effect on the milk FA composition in HF were shown to be much smaller in the breeds MRY, DF, and GWH, especially for the MRY breed.

Multibreed genomic prediction using multitrait genomic residual maximum likelihood and multitask Bayesian variable selection

“…Third, the traits analyzed were all associated with a few QTL of moderate to large effects in Holsteins. For those QTL with relatively large effects, the actual effects are likely to be similar in magnitude across breeds; for instance, as is shown regarding the effect of the DGAT1 gene (Spelman et al, 2002;Thaller et al, 2003;Maurice-Van Eijndhoven et al, 2015). In other words, for this specific group of QTL, the genetic correlation between breeds is expected to be close to unity if the same QTL segregate in both breeds.…”

“…The limited benefit of pooling the training data for fat yield, and the low estimate of the genetic correlation for fat yield, may be partly due to differences in estimated effects for Holstein and Jersey in the detected QTL regions on BTA 5 [associated with the MGST1 gene in Holsteins (Wang et al, 2012;Raven et al, 2014;Kemper et al, 2015a;Maurice-Van Eijndhoven et al, 2015;Littlejohn et al, 2016)] and on BTA 14 [associated with the DGAT1 gene, which has a strong effect on fat yield (Grisart et al, 2002;Boichard et al, 2003;Schennink et al, 2007)]. Possible explanations for the differences in estimated effects between Holstein and Jerseys for these regions are that the SNP on BTA 5 and 14 with the largest signal in Holstein had very low minor allele frequency in Jersey.…”

Conservation priorities for the different lines of Dutch Red and White Friesian cattle change when relationships with other breeds are taken into account

“…; Maurice‐Van Eijndhoven et al . ) . These data included 1287 purebred cows; 989 were Holstein Friesian (HF), 97 Groningen White headed (GWH), 137 Meuse‐Rhine‐Yssel (MRY) and 64 Dutch Friesian (DF).…”

“…After these editing steps, 51 974 of the initial 54 609 SNPs remained. Data from the DFR cattle were supplemented with data originating from studies with four other Dutch breeds (Maurice-van Eijndhoven 2014; Pryce et al 2014;Maurice-Van Eijndhoven et al 2015) . These data included 1287 purebred cows; 989 were Holstein Friesian (HF), 97 Groningen White headed (GWH), 137 Meuse-Rhine-Yssel (MRY) and 64 Dutch Friesian (DF).…”

Additive-Polygenic Inheritance of Reproductive System Diseases in Holstein Cows in Subpopulations

“…This group of diseases is characterized by multifactorial (due to many loci) control of stability and susceptibility under significant influence of environmental conditions (additive-polygenic inheritance [7]. When certain values of predisposition -the so-called threshold of susceptibility-are exceeded, the mechanism of multifactorial disease development is triggered [8].…”