Combining multi-population datasets for joint genome-wide association and meta-analyses: The case of bovine milk fat composition traits.

In genome-wide association studies (GWAS), sample size is the most important factor affecting statistical power that is under control of the investigator, posing a major challenge in understanding the genetics underlying difficult-to-measure traits. Combining data sets available from different populations for joint or meta-analysis is a promising alternative to increasing sample sizes available for GWAS. Simulation studies indicate statistical advantages from combining raw data or GWAS summaries in enhancing quantitative trait loci (QTL) detection power. However, the complexity of genetics underlying most quantitative traits, which itself is not fully understood, is difficult to fully capture in simulated data sets. In this study, population-specific and combined-population GWAS as well as a meta-analysis of the population-specific GWAS summaries were carried out with the objective of assessing the advantages and challenges of different data-combining strategies in enhancing detection power of GWAS using milk fatty acid (FA) traits as examples. Gas chromatography (GC) quantified milk FA samples and high-density (HD) genotypes were available from 1,566 Dutch, 614 Danish, and 700 Chinese Holstein Friesian cows. Using the joint GWAS, 28 additional genomic regions were detected, with significant associations to at least 1 FA, compared with the population-specific analyses. Some of these additional regions were also detected using the implemented meta-analysis. Furthermore, using the frequently reported variants of the diacylglycerol acyltransferase 1 (DGAT1) and stearoyl-CoA desaturase (SCD1) genes, we show that significant associations were established with more FA traits in the joint GWAS than the remaining scenarios. However, there were few regions detected in the population-specific analyses that were not detected using the joint GWAS or the meta-analyses. Our results show that combining multi-population data set can be a powerful tool to enhance detection power in GWAS for seldom-recorded traits. Detection of a higher number of regions using the meta-analysis, compared with any of the population-specific analyses also emphasizes the utility of these methods in the absence of raw multi-population data sets to undertake joint GWAS.

Gebreyesus G ，Buitenhuis AJ ，Poulsen NA ，Visker MHPW ，Zhang Q ，van Valenberg HJF ，Sun D ，Bovenhuis H ... -  《-》

Joint genome-wide association study for milk fatty acid traits in Chinese and Danish Holstein populations.

The identification of causal genes or genomic regions associated with fatty acids (FA) will enhance our understanding of the pathways underlying FA synthesis and provide opportunities for changing milk fat composition through a genetic approach. The linkage disequilibrium between adjacent markers is highly consistent between the Chinese and Danish Holstein populations, such that a joint genome-wide association study (GWAS) can be performed. In this study, a joint GWAS was performed for 16 milk FA traits based on data of 784 Chinese and 371 Danish Holstein cows genotyped by a high-density bovine single nucleotide polymorphism (SNP) array. A total of 486,464 SNP markers on 29 bovine autosomes were used. Bonferroni corrections were applied to adjust the significance thresholds for multiple testing at the genome- and chromosome-wide levels. According to the analysis of either the Chinese or Danish data individually, the total numbers of overlapping SNP that were significant at the chromosome level were 94 for C14:1, 208 for the C14 index, and 1 for C18:0. Joint analysis using the combined data of the 2 populations detected greater numbers of significant SNP compared with either of the individual populations alone for 7 and 10 traits at the genome- and chromosome-wide significance levels, respectively. Greater numbers of significant SNP were detected for C18:0 and the C18 index in the Chinese population compared with the joint analysis. Sixty-five significant SNP across all traits had significantly different effects in the 2 populations. Ten FA were influenced by a quantitative trait loci (QTL) region including DGAT1. Both C14:1 and the C14 index were influenced by a QTL region including SCD1 in the combined population. Other QTL regions also showed significant associations with the studied FA. A large region (14.9-24.9 Mbp) in BTA26 significantly influenced C14:1 and the C14 index in both populations, mostly likely due to the SNP in SCD1. A QTL region (69.97-73.69 Mbp) on BTA9 showed a significantly different effect on C18:0 between the 2 populations. Detection of these important SNP and the corresponding QTL regions will be helpful for follow-up studies to identify causal mutations and their interaction with environments for milk FA in dairy cattle.

Li X ，Buitenhuis AJ ，Lund MS ，Li C ，Sun D ，Zhang Q ，Poulsen NA ，Su G ... -  《-》

Genome-wide association and biological pathway analysis for milk-fat composition in Danish Holstein and Danish Jersey cattle.

The milk fat profile of the Danish Holstein (DH) and Danish Jersey (DJ) show clear differences. Identification of the genomic regions, genes and biological pathways underlying the milk fat biosynthesis will improve the understanding of the biology underlying bovine milk fat production and may provide new possibilities to change the milk fat composition by selective breeding. In this study a genome wide association scan (GWAS) in the DH and DJ was performed for a detailed milk fatty acid (FA) profile using the HD bovine SNP array and subsequently a biological pathway analysis based on the SNP data was performed. The GWAS identified in total 1,233 SNPs (FDR < 0.10) spread over 18 chromosomes for nine different FA traits for the DH breed and 1,122 SNPs (FDR < 0.10) spread over 26 chromosomes for 13 different FA traits were detected for the DJ breed. Of these significant SNPs, 108 SNP markers were significant in both DH and DJ (C14-index, BTA26; C16, BTA14; fat percentage (FP), BTA14). This was supported by an enrichment test. The QTL on BTA14 and BTA26 represented the known candidate genes DGAT and SCD. In addition we suggest ACSS3 to be a good candidate gene for the QTL on BTA5 for C10:0 and C15:0. In addition, genetic correlations between the FA traits within breed showed large similarity across breeds. Furthermore, the biological pathway analysis revealed that fat digestion and absorption (KEGG04975) plays a role for the traits FP, C14:1, C16 index and C16:1. There was a clear similarity between the underlying genetics of FA in the milk between DH and DJ. This was supported by the fact that there was substantial overlap between SNPs for FP, C14 index, C14:1, C16 index and C16:1. In addition genetic correlations between FA showed a similar pattern across DH and DJ. Furthermore the biological pathway analysis suggested that fat digestion and absorption KEGG04975 is important for the traits FP, C14:1, C16 index and C16:1.

Buitenhuis B ，Janss LL ，Poulsen NA ，Larsen LB ，Larsen MK ，Sørensen P ... -  《BMC GENOMICS》

Multi-population GWAS and enrichment analyses reveal novel genomic regions and promising candidate genes underlying bovine milk fatty acid composition.

The power of genome-wide association studies (GWAS) is often limited by the sample size available for the analysis. Milk fatty acid (FA) traits are scarcely recorded due to expensive and time-consuming analytical techniques. Combining multi-population datasets can enhance the power of GWAS enabling detection of genomic region explaining medium to low proportions of the genetic variation. GWAS often detect broader genomic regions containing several positional candidate genes making it difficult to untangle the causative candidates. Post-GWAS analyses with data on pathways, ontology and tissue-specific gene expression status might allow prioritization among positional candidate genes. Multi-population GWAS for 16 FA traits quantified using gas chromatography (GC) in sample populations of the Chinese, Danish and Dutch Holstein with high-density (HD) genotypes detects 56 genomic regions significantly associated to at least one of the studied FAs; some of which have not been previously reported. Pathways and gene ontology (GO) analyses suggest promising candidate genes on the novel regions including OSBPL6 and AGPS on Bos taurus autosome (BTA) 2, PRLH on BTA 3, SLC51B on BTA 10, ABCG5/8 on BTA 11 and ALG5 on BTA 12. Novel genes in previously known regions, such as FABP4 on BTA 14, APOA1/5/7 on BTA 15 and MGST2 on BTA 17, are also linked to important FA metabolic processes. Integration of multi-population GWAS and enrichment analyses enabled detection of several novel genomic regions, explaining relatively smaller fractions of the genetic variation, and revealed highly likely candidate genes underlying the effects. Detection of such regions and candidate genes will be crucial in understanding the complex genetic control of FA metabolism. The findings can also be used to augment genomic prediction models with regions collectively capturing most of the genetic variation in the milk FA traits.

Gebreyesus G ，Buitenhuis AJ ，Poulsen NA ，Visker MHPW ，Zhang Q ，van Valenberg HJF ，Sun D ，Bovenhuis H ... -  《BMC GENOMICS》

Reliability of genomic prediction for milk fatty acid composition by using a multi-population reference and incorporating GWAS results.

Large-scale phenotyping for detailed milk fatty acid (FA) composition is difficult due to expensive and time-consuming analytical techniques. Reliability of genomic prediction is often low for traits that are expensive/difficult to measure and for breeds with a small reference population size. An effective method to increase reference population size could be to combine datasets from different populations. Prediction models might also benefit from incorporation of information on the biological underpinnings of quantitative traits. Genome-wide association studies (GWAS) show that genomic regions on Bos taurus chromosomes (BTA) 14, 19 and 26 underlie substantial proportions of the genetic variation in milk FA traits. Genomic prediction models that incorporate such results could enable improved prediction accuracy in spite of limited reference population sizes. In this study, we combine gas chromatography quantified FA samples from the Chinese, Danish and Dutch Holstein populations and implement a genomic feature best linear unbiased prediction (GFBLUP) model that incorporates variants on BTA14, 19 and 26 as genomic features for which random genetic effects are estimated separately. Prediction reliabilities were compared to those estimated with traditional GBLUP models. Predictions using a multi-population reference and a traditional GBLUP model resulted in average gains in prediction reliability of 10% points in the Dutch, 8% points in the Danish and 1% point in the Chinese populations compared to predictions based on population-specific references. Compared to the traditional GBLUP, implementation of the GFBLUP model with a multi-population reference led to further increases in prediction reliability of up to 38% points in the Dutch, 23% points in the Danish and 13% points in the Chinese populations. Prediction reliabilities from the GFBLUP model were moderate to high across the FA traits analyzed. Our study shows that it is possible to predict genetic merits for milk FA traits with reasonable accuracy by combining related populations of a breed and using models that incorporate GWAS results. Our findings indicate that international collaborations that facilitate access to multi-population datasets could be highly beneficial to the implementation of genomic selection for detailed milk composition traits.

Gebreyesus G ，Bovenhuis H ，Lund MS ，Poulsen NA ，Sun D ，Buitenhuis B ... -  《-》