Christine Baes

Evaluation of variant identification methods for whole genome sequencing data in dairy cattle

BackgroundAdvances in human genomics have allowed unprecedented productivity in terms of algorithms, software, and literature available for translating raw next-generation sequence data into high-quality information. The challenges of variant identification in organisms with lower quality reference genomes are less well documented. We explored the consequences of commonly recommended preparatory steps and the effects of single and multi sample variant identification methods using four publicly available software applications (Platypus, HaplotypeCaller, Samtools and UnifiedGenotyper) on whole genome sequence data of 65 key ancestors of Swiss dairy cattle populations. Accuracy of calling next-generation sequence variants was assessed by comparison to the same loci from medium and high-density single nucleotide variant (SNV) arrays.ResultsThe total number of SNVs identified varied by software and method, with single (multi) sample results ranging from 17.7 to 22.0 (16.9 to 22.0) million variants. Computing time varied considerably between software. Preparatory realignment of insertions and deletions and subsequent base quality score recalibration had only minor effects on the number and quality of SNVs identified by different software, but increased computing time considerably. Average concordance for single (multi) sample results with high-density chip data was 58.3% (87.0%) and average genotype concordance in correctly identified SNVs was 99.2% (99.2%) across software. The average quality of SNVs identified, measured as the ratio of transitions to transversions, was higher using single sample methods than multi sample methods. A consensus approach using results of different software generally provided the highest variant quality in terms of transition/transversion ratio.ConclusionsOur findings serve as a reference for variant identification pipeline development in non-human organisms and help assess the implication of preparatory steps in next-generation sequencing pipelines for organisms with incomplete reference genomes (pipeline code is included). Benchmarking this information should prove particularly useful in processing next-generation sequencing data for use in genome-wide association studies and genomic selection.

Refined positioning of a quantitative trait locus affecting somatic cell score on chromosome 18 in the German Holstein using linkage disequilibrium

Combined linkage and linkage disequilibrium analysis (LALD) was conducted to more accurately map a previously reported quantitative trait locus (QTL) affecting somatic cell score on bovine chromosome 18. A grand-daughter design consisting of 6 German Holstein grandsire families with 1,054 progeny-tested genotyped sons was used in this study. Twenty microsatellite markers, 5 single nucleotide polymorphisms, and an erythrocyte antigen marker with an average marker spacing of 1.95 cM were analyzed along a chromosomal segment of 50.80 cM. Variance components were estimated and restricted maximum likelihood test statistics were calculated at the midpoint of each marker interval. The test statistics calculated in single-QTL linkage analysis exceeded the genome-wide significance threshold at several putative QTL positions. Using LALD, we were successful in assigning a genome-wide significant QTL to a confidence interval of 10.8 cM between the markers ILSTS002 and BMS833. The QTL in this marker interval was estimated to be responsible for between 5.89 and 13.86% of the genetic variation in somatic cell score. In contrast to the single-QTL linkage analysis model, LALD analyses with a 2-QTL model confirmed the position of one QTL, but gave no conclusive evidence for the existence or position of a second QTL. Ultimately, the QTL position was narrowed down considerably compared with previous results with a refined confidence interval of less than 11 cM.

Invited review: Inbreeding in the genomics era: Inbreeding, inbreeding depression, and management of genomic variability

Traditionally, pedigree-based relationship coefficients have been used to manage the inbreeding and degree of inbreeding depression that exists within a population. The widespread incorporation of genomic information in dairy cattle genetic evaluations allows for the opportunity to develop and implement methods to manage populations at the genomic level. As a result, the realized proportion of the genome that 2 individuals share can be more accurately estimated instead of using pedigree information to estimate the expected proportion of shared alleles. Furthermore, genomic information allows genome-wide relationship or inbreeding estimates to be augmented to characterize relationships for specific regions of the genome. Region-specific stretches can be used to more effectively manage areas of low genetic diversity or areas that, when homozygous, result in reduced performance across economically important traits. The use of region-specific metrics should allow breeders to more precisely manage the trade-off between the genetic value of the progeny and undesirable side effects associated with inbreeding. Methods tailored toward more effectively identifying regions affected by inbreeding and their associated use to manage the genome at the herd level, however, still need to be developed. We have reviewed topics related to inbreeding, measures of relatedness, genetic diversity and methods to manage populations at the genomic level, and we discuss future challenges related to managing populations through implementing genomic methods at the herd and population levels.

A 100-Year Review: Identification and genetic selection of economically important traits in dairy cattle

Over the past 100 yr, the range of traits considered for genetic selection in dairy cattle populations has progressed to meet the demands of both industry and society. At the turn of the 20th century, dairy farmers were interested in increasing milk production; however, a systematic strategy for selection was not available. Organized milk performance recording took shape, followed quickly by conformation scoring. Methodological advances in both genetic theory and statistics around the middle of the century, together with technological innovations in computing, paved the way for powerful multitrait analyses. As more sophisticated analytical techniques for traits were developed and incorporated into selection programs, production began to increase rapidly, and the wheels of genetic progress began to turn. By the end of the century, the focus of selection had moved away from being purely production oriented toward a more balanced breeding goal. This shift occurred partly due to increasing health and fertility issues and partly due to societal pressure and welfare concerns. Traits encompassing longevity, fertility, calving, health, and workability have now been integrated into selection indices. Current research focuses on fitness, health, welfare, milk quality, and environmental sustainability, underlying the concentrated emphasis on a more comprehensive breeding goal. In the future, on-farm sensors, data loggers, precision measurement techniques, and other technological aids will provide even more data for use in selection, and the difficulty will lie not in measuring phenotypes but rather in choosing which traits to select for.

Identification of a two-marker-haplotype on Bos taurus autosome 18 associated with somatic cell score in German Holstein cattle

BackgroundThe somatic cell score (SCS) is implemented in routine sire evaluations in many countries as an indicator trait for udder health. Somatic cell score is highly correlated with clinical mastitis, and in the German Holstein population quantitative trait loci (QTL) for SCS have been repeatedly mapped on Bos taurus autosome 18 (BTA18). In the present study, we report a refined analysis of previously detected QTL regions on BTA18 with the aim of identifying marker and marker haplotypes in linkage disequilibrium with SCS. A combined linkage and linkage disequilibrium approach was implemented, and association analyses of marker genotypes and maternally inherited two-marker-haplotypes were conducted to identify marker and haplotypes in linkage disequilibrium with a locus affecting SCS in the German Holstein population.ResultsWe detected a genome-wide significant QTL within marker interval 9 (HAMP_c.366+109G>A - BMS833) in the middle to telomeric region on BTA18 and a second putative QTL in marker interval 12-13 (BB710 - PVRL2_c.392G>A). Association analyses with genotypes of markers flanking the most likely QTL positions revealed the microsatellite marker BMS833 (interval 9) to be associated with a locus affecting SCS within the families investigated. A further analysis of maternally inherited two-marker haplotypes and effects of maternally inherited two-marker-interval gametes indicated haplotype 249-G in marker interval 12-13 (BB710 - PVRL2_c.392G>A) to be associated with SCS in the German Holstein population.ConclusionOur results confirmed previous QTL mapping results for SCS and support the hypothesis that more than one locus presumably affects udder health in the middle to telomeric region of BTA18. However, a subsequent investigation of the reported QTL regions is necessary to verify the two-QTL hypothesis and confirm the association of two-marker-haplotype 249-G in marker interval 12-13 (BB710 - PVRL2_c.392G>A) with SCS. For this purpose, higher marker density and multiple-trait and multiple-QTL models are required to narrow down the position of the causal mutation or mutations affecting SCS in German Holstein cattle.

Short communication: Genetic correlations between number of embryos produced using in vivo and in vitro techniques in heifer and cow donors

Multiple embryos can be produced from a heifer or cow donors using an in vivo or an in vitro technique. Comparisons of the number of embryos produced by the same donors as heifers and cows and using different techniques are limited. The main objectives of this study were to assess the genetic correlation between the number of embryos produced by Holstein donors using an in vivo and in vitro technique as a heifer and as a cow. The data set used was recorded by Holstein Canada and included all successful superovulations or ovum pickup and in vitro fertilization procedures performed on Holstein donors for more than 20yr. The type of technique used was known for all records and the status of the donor at recovery was retrieved from calving records. Bivariate repeatability animal model analyses were performed for both the total number of embryos (NE) and the number of viable embryos (VE) recovered per procedure. Logarithmic transformation was performed on the traits to normalize the data. Heritability estimates for the donor varied between 0.14 (0.02) and 0.19 (0.03) over all analyses, indicating that the number of embryos produced by a donor is influenced by the genetic potential of the donor. Genetic correlations between records produced in vivo and in vitro were moderately high and positive (NE=0.85±0.07; VE=0.63±0.09), suggesting that donors with high genetic potential for in vivo superovulation tend also to have high potential to produce multiple embryos in vitro. Similarly, the moderately high genetic correlations (NE=0.79±0.05; VE=0.72±0.05) found between heifer and cow records indicate that a donor tends to produce a comparable number of embryos as a heifer or as a cow. The estimated repeatabilities (0.23 to 0.35) indicated that the number of embryos recovered should be somewhat repeatable in the same donor over time. On the other hand, the service sires seem not to play an important role on the total number of embryos produced by a donor no matter the technique used or the status of the donor at recovery.

Genome-wide association study and in silico functional analysis of the number of embryos produced by Holstein donors

Superovulation or ovum pick-up and in vitro fertilization are technologies used to produce an increased number of embryos from elite females. Embryo production traits have been shown to be heritable, but the genes that cause this variability have not yet been assessed. The main objectives of this study were to perform a genome-wide association study (GWAS) to find single nucleotide polymorphisms (SNP) associated with embryo production traits and to identify candidate genes affecting the number of embryos produced by Holstein donors in Canada that may provide insight into the regulation of embryo production. Breeding values were estimated and de-regressed for all donors and sires using a data set of 150,971 records of superovulation or ovum pick-up and in vitro fertilization. A total of 11,607 animals were genotyped, but of that number only 5,118 were genotyped with at least a 50K SNP panel and had a de-regressed estimated breeding value reliability of at least 10%. For the GWAS, 606,406 imputed SNP on 29 autosomal chromosomes were considered after applying quality control measures. A single-SNP univariate mixed linear animal model was used to perform the GWAS, and a 5% false discovery rate was applied to adjust for multiple testing. We found 36 and 14 significant SNP associated with the total number of embryos and the number of viable embryos, respectively, with most of them located on chromosome 11. Using these significant SNP, positional genes located within 10,000 bp upstream and downstream of the SNP were retrieved. Thirteen genes were harboring or near the significant SNP for the total number of embryos, 4 of them also being near the significant SNP for viable embryos. Some of these genes (CRB2, DENND1A, MAD1L1, NDUFA8, PTGS1) could be considered as potential positional candidate genes related to the number of embryos produced by a donor. This list will need to be validated in an independent population to confirm the role of the genes for embryo production.

Genetic analysis for quality of frozen embryos produced by Holstein cattle donors in Canada

The number of embryos produced by Holstein donors has been shown to be heritable, so it could be possible to genetically select for this trait to improve the efficiency of the assisted reproductive technology (ART) in dairy cattle. Another important parameter to consider for achieving good results from ART is embryo quality because embryos of good quality have more chance of producing live offspring. The possibility of using genetic selection for increasing the quality of embryo produced from ART has yet to be assessed. The objective of this study was, therefore, to perform a genetic analysis of embryo quality of Holstein donors in Canada using data recorded by Holstein Canada. The data set used was missing quality score data for embryos transferred fresh into a recipient, so the analyses were only performed for frozen embryos. With most traits in the Canadian dairy industry being evaluated with linear models, embryo quality was also evaluated with this class of models. However, considering the categorical nature of embryo quality, a threshold model was also evaluated. Embryo quality data were analyzed with either a univariate linear animal model or a univariate binomial threshold animal model. Genetic parameters estimated from the different models were comparable. A low heritability was found for the donor (0.04 ± <0.01) and the service sire (0.02 ± <0.01), but the repeatability estimate for the donor was higher (0.17), indicating that it was worthwhile to use a repeated records model. Overall, considering the low genetic parameters estimated, slow genetic progress is expected for the quality of frozen embryos produced by Canadian Holstein donors. Rank correlations were calculated between breeding values estimated from different models. High correlations were found between all models, indicating that no substantial re-ranking of the animals is expected from the different models. So, even though a threshold model is better suited for the analysis of categorical data, a linear model could be used for the analysis of embryo quality because it is less computationally demanding.

Comparison of genomic predictions for lowly heritable traits using multi-step and single-step genomic best linear unbiased predictor in Holstein cattle

The success and sustainability of a breeding program incorporating genomic information is largely dependent on the accuracy of predictions. For low heritability traits, large training populations are required to achieve high accuracies of genomic estimated breeding values (GEBV). By including genotyped and nongenotyped animals simultaneously in the evaluation, the single-step genomic BLUP (ssGBLUP) approach has the potential to deliver more accurate and less biased genomic evaluations. The aim of this study was to compare the accuracy and bias of genomic predictions for various traits in Canadian Holstein cattle using ssGBLUP and multi-step genomic BLUP (msGBLUP) under different strategies, such as (1) adding genomic information of cows in the analysis, (2) testing different adjustments of the genomic relationship matrix, and (3) using a blending approach to obtain GEBV from msGBLUP. The following genomic predictions were evaluated regarding accuracy and bias: (1) GEBV estimated by ssGBLUP; (2) direct genomic value estimated by msGBLUP with polygenic effects of 5 and 20%; and (3) GEBV calculated by a blending approach of direct genomic value with estimated breeding values using polygenic effects of 5 and 20%. The effect of adding genomic information of cows in the evaluation was also assessed for each approach. When genomic information was included in the analyses, the average improvement in observed reliability of predictions was observed to be 7 and 13 percentage points for reproductive and workability traits, respectively, compared with traditional BLUP. Absolute deviation from 1 of the regression coefficient of the linear regression of de-regressed estimated breeding values on genomic predictions went from 0.19 when using traditional BLUP to 0.22 when using the msGBLUP method, and to 0.14 when using the ssGBLUP method. The use of polygenic weight of 20% in the msGBLUP slightly improved the reliability of predictions, while reducing the bias. A similar trend was observed when a blending approach was used. Adding genomic information of cows increased reliabilities, while decreasing bias of genomic predictions when using the ssGBLUP method. Differences between using a training population with cows and bulls or with only bulls for the msGBLUP method were small, likely due to the small number of cows included in the analysis. Predictions for lowly heritable traits benefit greatly from genomic information, especially when all phenotypes, pedigrees, and genotypes are used in a single-step approach.

Candidate gene association analyses for ketosis resistance in Holsteins

High-yielding dairy cattle are susceptible to ketosis, a metabolic disease that negatively affects the health, fertility, and milk production of the cow. Interest in breeding for more robust dairy cattle with improved resistance to disease is global; however, genetic evaluations for ketosis would benefit from the additional information provided by genetic markers. Candidate genes that are proposed to have a biological role in the pathogenesis of ketosis were investigated in silico and a custom panel of 998 putative single nucleotide polymorphism (SNP) markers was developed. The objective of this study was to test the associations of these new markers with deregressed estimated breeding values (EBV) for ketosis. A sample of 653 Canadian Holstein cows that had been previously genotyped with a medium-density SNP chip were regenotyped with the custom panel. The EBV for ketosis in first and later lactations were obtained for each animal and deregressed for use as pseudo-phenotypes for association analyses. Results of the mixed inheritance model for single SNP association analyses suggested 15 markers in 6 unique candidate genes were associated with the studied trait. Genes encoding proteins involved in metabolic processes, including the synthesis and degradation of fatty acids and ketone bodies, gluconeogenesis, lipid mobilization, and the citric acid cycle, were identified to contain SNP associated with ketosis resistance. This work confirmed the presence of previously described quantitative trait loci for dairy cattle, suggested novel markers for ketosis-resistance, and provided insight into the underlying biology of this disease.