Associated effects of copy number variants on economically important traits in Spanish Holstein dairy cattle

Abstract

Copy number variants (CNV) are structural variants consisting of duplications or deletions of genomic fragments longer than 1 kb that present variability in the population and are heritable. The objective of this study was to identify CNV regions (CNVR) associated with 7 economically important traits (production, functional, and type traits) in Holstein cattle: fat yield, protein yield, somatic cell count, days open, stature, foot angle, and udder depth. Copy number variants were detected by using deep-sequencing data from 10 sequenced bulls and the Bovine SNP chip array hybridization signals. To reduce the number of false-positive calls, only CNV identified by both sequencing and Bovine SNP chip assays were kept in the final data set. This resulted in 823 CNVR. After filtering by minor allele frequency >0.01, a total of 90 CNVR appeared segregating in the bulls that had phenotypic data. Linear and quadratic CNVR effects were estimated using Bayesian approaches. A total of 15 CNVR were associated with the traits included in the analysis. One CNVR was associated with fat and protein yield, another 1 with fat yield, 3 with stature, 1 with foot angle, 7 with udder depth, and only 1 with days open. Among the genes located within these regions, highlighted were the MTHFSD gene that belongs to the folate metabolism genes, which play critical roles in regulating milk protein synthesis; the SNRPE gene that is related to several morphological pathologies; and the NF1 gene, which is associated with potential effects on fertility traits. The results obtained in the current study revealed that these CNVR segregate in the Holstein population, and therefore some potential exists to increase the frequencies of the favorable alleles in the population after independent validation of results in this study. However, genetic variance explained by the variants reported in this study was small.

CNV discovery for milk composition traits in dairy cattle using whole genome resequencing

“…After converting 6015 CNVRs to corresponding results based on Btau4.0 using the UCSC liftOver tool with 50% of bases that must remap, 3996 CNVRs of which were successfully converted amounting to about 45.06 Mb. We found that ~80% of the 3996 CNVRs overlapped with those reported by previous investigations [23, 27–30] by 1 bp or greater (Fig. 5), and the largest overlap was ~7.92 Mb of the reported by Bickhart et al [23].…”

“…The number of CNVs (14,821) identified in this study was more than the reports based on NGS data by Bichkhart et al (1265) in Angus, Holstein, Hereford and Nelore cattle [23], Stothard et al (790) in Holstein and Black Angus [28], Zhan et al (520) in Holstein [27], Boussaha et al (957) in Holstein, Montbéliarde and Normande [29] and Ben et al (823) in Holstein [30]. In addition, Jiang et al detected CNVs based on Illumina BovineSNP50 (99) and BovineHD chips (367) data in Chinese Holstein population [24, 25], which were less than what we detected in this study.…”

“…As for the above inconsistencies, there are likely due to different detection methods and different samples. Bickhart et al used mrFAST/mrsFAST and WSSD [23], and both Zhan et al [27] and Stothard et al [28] performed CNV-seq, and Boussaha et al [29] used GATK, Pindel, and Ben et al [30] performed control-FREE. While in this study, we used CNVnator.…”

“…For the CNV detection in the cattle genome, there have been several studies reported using such methods, including CGH [67, 68], BovineSNP50 Beadchip [32, 69, 70], BovineHD SNP Beadchip [25, 31] and NGS [23, 27–30]. In this study, the objective was to identify candidate genes for milk protein and fat traits of dairy cattle through CNV detection based on NGS data of specific Holstein bulls that have extremely high and low estimated breeding values (EBVs) for milk protein and fat percentages.…”

Evolutionary and Functional Features of Copy Number Variation in the Cattle Genome1

“…Cattle CNVs have been reported using a variety of platforms, including comparative genomic hybridization arrays (Liu et al, 2008, 2010; Fadista et al, 2010), the Illumina BovineHD BeadChip (Hou et al, 2012a; Wu et al, 2015; Aguilar et al, 2016; Prinsen et al, 2016; Xu et al, 2016), the Illumina BovineSNP50 BeadChip (Matukumalli et al, 2009; Bae et al, 2010; Hou et al, 2011, 2012b; Jiang et al, 2012; Bagnato et al, 2015; Ben Sassi et al, 2016), and next-generation sequencing (NGS) (Stothard et al, 2011; Zhan et al, 2011; Bickhart et al, 2012; Choi et al, 2013; Keel et al, 2016; Ben Sassi et al, 2016). In these studies, it is reported that copy number variable regions comprise ~2–7% of the cattle genome.…”