Identification of candidate markers on bovine chromosome 14 (BTA14) under milk production trait quantitative trait loci in Holstein

The objective of this study was to identify single-nucleotide polymorphisms using a bovine chromosome 14 high-density SNP panel after accounting for the effect of DGAT1. Linkage disequilibrium information and sire heterozygosity were used to select markers for linkage analysis on bovine chromosome 14 for milk production traits in 321 Holstein animals. Results show putative milk peaks at 42 and 61 cM, both at p<0.10, a fat yield peak at 42 and 63 cM, both at p<0.05; a protein yield peak at 42 (p<0.01) and 84 cM (p<0.05); fat per cent peaks at 3 (p<0.01) and 29 cM (p<0.05), and a protein per cent peak at 4 cM (p<0.05). Once quantitative trait loci positions were established, allele substitution effects for all markers were evaluated using the same statistical model. Overlaying information between quantitative trait loci (QTL) and allele effect analysis enabled the identification (p<0.01) of 20 SNPs under the milk yield QTL, 2 under both of the fat yield peaks, 8 and 9 under the protein yield peaks, 2 and 6 for the fat per cent peaks and 5 for the protein per cent peak. One SNP in particular, ss61514555:A>C, showed association with 3 of the 5 traits: milk (p=1.59E-04), fat (p=6.88E-05) and protein yields (p=5.76E-05). Overall, combining information from linkage disequilibrium, sire heterozygosity and genetic knowledge of traits enabled the characterization of additional markers with significant associations with milk production traits.     

Refined mapping of twinning-rate quantitative trait loci on bovine chromosome 5 and analysis of insulin-like growth factor-1 as a positional candidate gene

Twinning in cattle is a complex trait that is associated with economic loss and health issues such as abortion, dystocia, and reduced calf survival. Twinning-rate QTL have been detected previously on BTA5 in the North American Holstein and Norwegian dairy cattle populations and in a USDA herd selected for high twinning rate. In previous work with the North American Holstein population, the strongest evidence for a QTL was obtained from analysis of an extended, multiple-generation family. Using additional animals, an increased density of SNP marker association tests, and a combined linkage and linkage disequilibrium mapping method, we refined the position of this QTL in the North American Holstein population. Two sets of twinning-rate predicted transmitting abilities estimated during 2 different time periods in the North American dairy cattle population were used to provide validation of results. A total of 106 SNP and 3 microsatellites were used to scan the genomic region between 5 and 80 Mb on BTA5. Combined linkage-linkage disequilibrium analysis identified significant evidence for QTL within the 25- to 35-Mb and 64- to 70-Mb regions of BTA5. The IGF-1 gene (IGF1) was examined as a positional candidate gene and an SNP in intron 2 of IGF1 was significantly associated with twinning rate by using both data sets (P = 0.003 and P = 1.05 x 10(-6)). Replication of this association in other cattle populations will be required to examine the extent of linkage disequilibrium with the underlying quantitative trait nucleotide across breeds.     

Genome-wide scans for QTL affecting carcass traits in Hereford x composite double backcross populations

A genome-wide scan for chromosomal regions influencing carcass traits was conducted spanning 2.413 morgans on 29 bovine autosomes using 229 microsatellite markers. Two paternal half-sib families of backcross progenies were produced by mating Hereford x composite gene combination (CGC) bulls to both Hereford and CGC dams. Progeny of the first sire (n = 146) were born in 1996 and progeny of the second sire (n = 112) were born in 1997. Each year cattle were fed out and slaughtered serially when they were between 614 and 741 d of age. Phenotypes measured at harvest were: live weight; carcass weight; fat depth; marbling; percentage kidney, pelvic, and heart fat (KPH); and rib eye area. Dressing percentage and USDA Yield Grade were calculated from these data. The phenotypes were adjusted to age-, live weight-, and fat depth-constant endpoints using analysis of covariance. The resulting residuals were analyzed by interval mapping to detect QTL. Within family, nominal significance was established by permutation analysis. Approximate genomewide significance levels were established by applying the Bonferroni correction to the nominal probability levels. Regression and error sums of squares and degrees of freedom were pooled across families when suggestive linkage identified in one family was confirmed in the other. The results indicate promising locations for QTL affecting live weight on BTA 17 and marbling on BTA 2 that segregate in Bos taurus. Also, previously identified linkage between central markers on BTA 5 and USDA Yield Grade was confirmed in one family. Greater marker saturation in these regions coupled with refined methods for data analysis will lead to more precise determination of QTL positions.     

Detection of quantitative trait loci affecting non-return rate in French dairy cattle

The purpose of this study was to map quantitative trait loci (QTL) influencing female fertility estimated by non-return rate (NRR) in the French dairy cattle breeds Prim'Holstein, Normande and Montbeliarde. The first step was a QTL detection study on NRR at 281 days after artificial insemination on 78 half-sib families including 4993 progeny tested bulls. In Prim'Holstein, three QTL were identified on Bos taurus chromosomes BTA01, BTA02 and BTA03 (p < 0.01), whereas one QTL was identified in Normande on BTA01 (p < 0.05). The second step aimed at confirming these three QTL and refining their location by selecting and genotyping additional microsatellite markers on a sub-sample of 41 families from the three breeds using NRR within 56, 90 and 281 days after AI. Only the three QTL initially detected in Prim'Holstein were confirmed. Moreover, the analysis of NRR within 56, 90 and 281 days after AI allowed us to distinguish two FF QTL on BTA02 in Prim'Holstein, one for NRR56 and one for NRR90. Estimated QTL variance was 18%, 14%, 11.5% and 14% of the total genetic variance, respectively, for QTL mapping to BTA01, BTA02 (NRR90 and NRR56) and BTA03.     

Variance component analysis of quantitative trait loci for pork carcass composition and meat quality on SSC4 and SSC11

In a previous study, QTL for carcass composition and meat quality were identified in a commercial finisher cross. The main objective of the current study was to confirm and fine map the QTL on SSC4 and SSC11 by genotyping an increased number of individuals and markers and to analyze the data using a combined linkage and linkage disequilibrium analysis method. A modified version of the method excludes linkage disequilibrium information from the analysis, enabling the comparison of results based on linkage information only or results based on combined linkage and linkage disequilibrium information. Nine additional paternal half-sib families were genotyped for 18 markers, resulting in a total of 1,855 animals genotyped for 15 and 13 markers on SSC4 and SSC11, respectively. The QTL affecting meat color on SSC4 was confirmed, whereas the QTL affecting LM weight could not be confirmed. The combined linkage and linkage disequilibrium analysis resulted in the identification of new significant effects for 14 traits on the 2 chromosomes. Heritabilities of the QTL effects ranged from 1.8 to 13.2%. The analysis contributed to a more accurate positioning of QTL and further characterized their phenotypic effect. However, results showed that even greater marker densities are required to take full advantage of linkage disequilibrium information and to identify haplotypes associated with favorable QTL alleles.     

Fine mapping quantitative trait loci under selective phenotyping strategies based on linkage and linkage disequilibrium criteria

In fine mapping of a large‐scale experimental population where collection of phenotypes are very expensive, difficult to record or time‐demanding, selective phenotyping could be used to phenotype the most informative individuals. Linkage analyses based sampling criteria (LAC) and linkage disequilibrium‐based sampling criteria (LDC) for selecting individuals to phenotype are compared to random phenotyping in a quantitative trait loci (QTL) verification experiment using stochastic simulation. Several strategies based on LAC and LDC for selecting the most informative 30%, 40% or 50% of individuals for phenotyping to extract maximum power and precision in a QTL fine mapping experiment were developed and assessed. Linkage analyses for the mapping was performed for individuals sampled on LAC within families and combined linkage disequilibrium and linkage analyses was performed for individuals sampled across the whole population based on LDC. The results showed that selecting individuals with similar haplotypes to the paternal haplotypes (minimum recombination criterion) using LAC compared to random phenotyping gave at least the same power to detect a QTL but decreased the accuracy of the QTL position. However, in order to estimate unbiased QTL parameters based on LAC in a large half‐sib family, prior information on QTL position was required. The LDC improved the accuracy to estimate the QTL position but not significantly compared to random phenotyping with the same sample size. When applying LDC (all phenotyping levels), the estimated QTL effect were closer to the true value in comparison to LAC. The results showed that the LDC were better than the LAC to select individuals for phenotyping and contributed to detection of the QTL.     

Adjustment of non-return rates for AI technicians and dairy bulls

Insemination results of technicians and dairy bulls in AI were studied using 87112 first inseminations from 283 bulls by 37 technicians. The traits studied were 28- and 56-day non-return rates. Models were compared on the basis of the correlation between solutions obtained from a reference model and reduced models.Age at first service in heifers and calving to first service interval in cows did not affect the ranking of sires and technicians. Routine monthly evaluation of non-return rates of sires required a correction for parity of cow and month of insemination. In addition genetic group and number of inseminations should be considered. Monthly evaluation of technicians could be based on deviations within studs.Annual evaluation of sires and technicians should consider a herd effect as well. The correlation between solutions (56-day non-return) for models with and without herd effect was 0.94 for both sires and technicians, whereas the absolute average change was 0.5%.     

Direct and maternal genetic parameters of fertility traits in Friesian cattle

Variance and covariance components for sire of cow and sire of insemination of fertility traits were estimated by parity using REML procedures. Subsequently direct genetic and maternal components were calculated. Heritabilities for 56-day non-return rate, conception rate (heifers, parity one to three) and days open (parity one to three) varied between 0.007 and 0.049. The heritability for age at conception in heifers was 0.192. The maternal component for days open was greater than the direct genetic component. The genetic correlations between sire of insemination and sire of cow for non-return rate changed from −0.29 in heifers to 0.59 in parity three, that for conception rate from −0.74 to 0.48. The relation between maternal and direct genetic components changed from −0.94 to 0.26 and −0.97 to 0.12 for non-return and conception rates, respectively. Changes in correlations for days open from parity one to three were smaller. The results indicated that the genetic correlation between the sire of cow components for non-return and conception rate in heifers, first parity and older cows may differ from unity.     

Evaluation of association between polymorphism within the thyroglobulin gene and milk production traits in dairy cattle

In dairy cattle, many studies have reported quantitative trait loci (QTL) on the centromeric end of chromosome 14 that affect milk production traits. One of the candidate genes in this QTL region – thyroglobulin (TG) – was previously found to be significantly associated with marbling in beef cattle. Thus, based on QTL studies in dairy cattle and because of possible effects of this gene on fat metabolism, we investigated the association of TG with milk yield and composition in Holstein dairy cattle. A total of 1279 bulls from the Cooperative Dairy DNA Repository Holstein population were genotyped for a single nucleotide polymorphism in TG used previously in beef cattle studies. Analysis of 29 sire families showed no significant association between TG variants and milk production traits. Within‐sire family analysis suggests that TG is neither the responsible gene nor a genetic marker in association with milk production traits.     

Refined mapping of quantitative trait loci for somatic cell score on BTA02 in the German Holstein

The aim of this study was to more precisely map a previously reported quantitative trait locus (QTL) affecting somatic cell score on Bos taurus autosome 2 by increasing the number of markers fourfold, analysing more families and exploiting within‐population linkage disequilibrium (LD). A granddaughter design of 10 German Holstein grandsire families with 1121 progeny tested sons was used. Twenty‐six markers with an average marker spacing of 3.14 cM were genotyped along 81.6 cM. Linkage analysis (LA) was performed using variance‐component methodology. The incorporation of LD was first done using variance‐component methods followed by regression on marker alleles. LA revealed genome‐wide significance (LOD > 3) at 15 contiguous marker‐intervals, with the maximum test‐statistic between DIK2862 and BMS778 and a 1‐lod drop‐off interval of 38 cM. While the variance‐component methods could not detect any LD, two individual markers with a significant effect (ILSTS098, p < 0.05; BMS778, p < 0.01) were found by regression analysis. Compared with previous results QTL‐localisation was substantially narrowed; further fine‐mapping should focus on the close vicinity of BMS778.     

An independent confirmation of a quantitative trait locus for milk yield and composition traits on bovine chromosome 26

Several reports have demonstrated that bovine chromosome 26 (BTA26) harbours significant or suggestive quantitative trait loci (QTL) for milk production and composition traits in dairy cattle. Our previous study showed that a C/T substitution in the bovine TCF7L2 gene on BTA26 was significantly linked to QTL for protein yield (PY) in a Canadian dairy cattle population. Actually, this polymorphism was one of the markers derived from a genome‐wide screening of QTL for milk PY using an amplified fragment length polymorphism technique combined with a DNA pooling strategy. In the present study, 990 Holstein bulls with complete genotype and phenotype data from 14 sire families were analysed to confirm, if the QTL effects exist in other populations. Statistical analysis revealed that this marker was significantly associated with PY, protein percentage, milk yield and fat yield (FY) (p < 0.001) in the US Holstein population. These results indicate that this QTL region has a pleiotrophic effect on different milk traits and is portable in different populations.     

Fine mapping and detection of the causative mutation underlying Quantitative Trait Loci

The effect on power and precision of including the causative SNP amongst the investigated markers in Quantitative Trait Loci (QTL) mapping experiments was investigated. Three fine mapping methods were tested to see which was most efficient in finding the causative mutation: combined linkage and linkage disequilibrium mapping (LLD); association mapping (MARK); a combination of LLD and association mapping (LLDMARK). Two simulated data sets were analysed: in one set, the causative SNP was included amongst the markers, while in the other set the causative SNP was masked between markers. Including the causative SNP amongst the markers increased both precision and power in the analyses. For the LLD method the number of correctly positioned QTL increased from 17 for the analysis without the causative SNP to 77 for the analysis including the causative SNP. The likelihood of the data analysis increased from 3.4 to 13.3 likelihood units for the MARK method when the causative SNP was included. When the causative SNP was masked between the analysed markers, the LLD method was most efficient in detecting the correct QTL position, while the MARK method was most efficient when the causative SNP was included as a marker in the analysis. The LLDMARK method, combining association mapping and LLD, assumes a QTL as the null hypothesis (using LLD method) and tests whether the ‘putative causative SNP’ explains significantly more variance than a QTL in the region. Thus, if the putative causative SNP does not only give an Identical‐By‐Descent (IBD) signal, but also an Alike‐In‐State (AIS) signal, LLDMARK gives a positive likelihood ratio. LLDMARK detected less than half as many causative SNPs as the other methods, and also had a relatively high false discovery rate when the QTL effect was large. LLDMARK may however be more robust against spurious associations, because the regional IBD is largely corrected for by fitting a QTL effect in the null hypothesis model.     

Estimation of linkage disequilibrium in a sample of the United Kingdom dairy cattle population using unphased genotypes

The association between genetic marker alleles was estimated for two regions of the bovine genome from a random sample of 50 young dairy bulls born in the United Kingdom between 1988 and 1995. Microsatellite marker genotypes were obtained for six markers on chromosome 2 and seven markers on chromosome 6, spanning 38 and 20 cM, respectively. Two different methods, which do not require family information, were used to estimate population haplotype frequencies. Haplotype frequencies were estimated for pairs of loci using the expectation-maximization algorithm and for all linked loci using a Bayesian approach via a Markov chain-Monte Carlo algorithm. Significant (P = 0.0007) linkage disequilibrium was detected between pairs of loci in syntenic groups (that is, loci in the same linkage group), extending to about 10 cM. No significant linkage disequilibrium was detected between markers in nonsyntenic regions. Given the observed level of linkage disequilibrium, mapping methods based on population-wide association might provide a better resolution than traditional quantitative trait loci mapping methods in the U.K. dairy cattle population and may reduce the required sample sizes of the experiments.     

Combined use of phenotypic and genotypic information in sampling animals for genotyping in detection of quantitative trait loci.

Conventional selective genotyping which is using the extreme phenotypes (EP) was compared with alternative criteria to find the most informative animals for genotyping with respects to mapping quantitative trait loci (QTL). Alternative sampling strategies were based on minimizing the sampling error of the estimated QTL effect (MinERR) and maximizing likelihood ratio test (MaxLRT) using both phenotypic and genotypic information. In comparison, animals were randomly genotyped either within or across families. One hundred data sets were simulated each with 30 half-sib families and 120 daughters per family. The strategies were compared in these datasets with respect to estimated effect and position of a QTL within a previously defined genomic region at genotyping 10, 20 or 30% of the animals. Combined linkage disequilibrium linkage analysis (LDLA) was applied in a variance component approach. Power to detect QTL was significantly higher for both MinERR and MaxLRT compared with EP and random genotyping methods (either across or within family), for all the proportions of genotyped animals. Power to detect significant QTL (alpha = 0.01) with 20% genotyping for MinERR and MaxLRT was 80 and 75% of that obtained with complete genotyping compared with 70 and 38% genotyping for EP within and across families respectively. With 30% genotyping, the powers were 78, 83, 78 and 58% respectively. The estimated variance components were unbiased in EP strategies (within and across family), only when at least 30% was genotyped. To decrease the number of genotyped individuals either MinERR or MaxLRT could be considered. With 20% genotyping in MinERR, the estimated QTL variance components were not significant compared with complete genotype information but all studied strategies at 20% genotyping overestimated the QTL effect. Results showed that combining the phenotypic and genotypic information in selective genotyping (e.g. MinERR and MaxLRT) is better than using only the EPs and the combined methods can be considered as alternative approaches to decrease genotyping costs, with unbiased QTL effects, decreased sampling variance of the QTL variance component and also increased the power of QTL detection.     

Estimation of genome-wide haplotype effects in half-sib designs

Genome-wide estimated breeding values can be computed from the simultaneous estimates of the effects of small intervals of DNA throughout the genome on a trait or traits of interest. Small intervals or segments of DNA can be created by the use of thousands of single nucleotide polymorphisms (SNP) available in panels of 10, 25 and 50 thousand SNP. A simulation study was conducted to compare factors that could influence the accuracy of genome-wide selection. Factors studied were the heritability of the trait, dispersion of quantitative trait loci (QTL) across the genome and size of the QTL effects. A 100-cM genome was assumed with 100 equally spaced SNP markers and 10 QTL. A granddaughter design was constructed with 20 sires and 100 sons per sire. Population-wide linkage disequilibrium was assumed to be sufficient after 25 generations of random mating starting with 30 sires and 400 dams. Best linear unbiased prediction was used to simultaneously estimate the effects of 99 SNP intervals, based on determining the SNP haplotype of each son inherited from the sire. Indicator variables were used in the model to indicate haplotype transmission. A genome-wide estimated breeding value was calculated as the sum of the appropriate haplotype interval estimates for each son. Correlations between estimated and true breeding values ranged from 0.60 to 0.79. Situations with unequally sized QTL effects and randomly dispersed QTL gave higher correlations. QTL positions could be estimated to within 2 cM or less.     

Accounting for heterogeneity of variances to improve the precision of QTL mapping in dairy cattle

The principle of interval mapping for quantitative trait loci (QTL) was originally developed for the analysis of single backcross data but it has been increasingly applied to more complicated experimental designs and data structures. It is important to study whether accounting for the heterogeneity of variance would improve the precision of QTL mapping based on data of multiple populations or families. This study compared homogeneous and heterogeneous maximum likelihood approaches for QTL mapping. The data consisted of 433 sons from six sire families with 69 microsatellite markers distributed over 12 chromosomes. The results of this study indicate that the heterogeneous approach generally produced a smaller residual variance and thus provided a better fit to the data than the homogeneous approach, meaning that the heterogeneous approach offers better precision in estimating both positions and effects of QTL. The results further showed that accounting for the heterogeneity of residual variance led to different statistical inferences from ignoring the heterogeneity of variance in QTL mapping. The heterogeneous approach is useful for QTL mapping based on the joint data of diverse reference populations or heteroscedastic data obtained from crossing animals with different genetic backgrounds.     

Initial results of genomic scans for ovulation rate in a cattle population selected for increased twinning rate

Genomic scans were conducted with 273 markers on 181 sires from a cattle population selected for increased twinning rate to identify chromosomal regions containing genes that influence ovulation rate. Criteria used for selecting markers were number of alleles, ease of scoring, and relative position within linkage group. Markers were multiplexed or multiple-loaded on the gels to reduce the costs and labor required to obtain genotypic data. This approach reduced the number of gels by 45% when compared with running each marker independently. Male animals selected for the genomic scan sired the majority of the population. A modified interval analysis was used in a granddaughter design to compare effects of each allele within sire for 10 different sire families. The midparent deviation of the son's estimated breeding value for ovulation rate was used as the phenotype. Forty-one potential peaks were identified with a nominal significance level < or = 0.05. The 10 peaks with the highest significance levels (P < 0.02) were selected for further analysis. Markers were genotyped across daughters of the sire where nominal significance was found for each of the 10 peaks. One peak (BTA5, relative position 40 cM) was found to be nominally significant in the daughters. The nominal significance levels were P = 0.01 for the sons (n = 32) and P = 0.02 for the daughters (n = 94) of sire 784403. A combined genomewide significance value (P = 0.07) was calculated that accounted for the 10 analyses with sons and the 10 analyses with daughters. These results strongly suggest that this region contains a gene(s) that is involved in the follicular recruitment and development process.     

Three types of acrosomal aberrations of bull spermatozoa and their relation to fertility

The effect of acrosomal aberrations of the spermatozoa of Finnish Ayrshire bulls on the corrected non-return rate within 60 days of the first 500 inseminations was studied. The material consisted of sperm samples examined by the artificial insemination societies. All samples had been accepted for use in artificial insemination. One Giemsa-stained slide was studied for each of the 95 bulls concerned. Samples showing distinct acrosomal defects were studied by electron microscopy. Three different types of acrosomal aberration were found. One was obviously associated with subfertility in all 6 bulls in which it was detected.     

Fine mapping of a QTL for intramuscular fat on porcine chromosome 6 using combined linkage and linkage disequilibrium mapping

In this study data from a commercial Norwegian slaughter pig cross was analysed to confirm a previous reported quantitative trait locus (QTL) affecting intramuscular fat (IMF) on porcine chromosome 6. The data consisted of an old experiment, in which the QTL was previously detected, and new experimental data from the Norwegian slaughter pig cross. The old and new experimental data were analysed separately and together. A previously described method combining linkage and linkage disequilibrium analysis (LDLA) was used for the analysis, but this method assumes that all animals are descendants from a common base population, which is not realistic in a cross between different breeds. An adjusted version of the method, able to distinguish between different breeds in the cross, is presented here. Using the LDLA method, we were not able to confirm the QTL in the old experimental data, because the genetic variance could be explained by the polygenic effect. Analysis from the new experimental data did however detect the QTL, and analysing the data from both experiments together gave highly significant results for a QTL (p < 0.001) between markers SW1355 and SW1823. The main conclusion is therefore that the previously reported QTL for IMF on porcine chromosome 6 was confirmed within a 8.7‐cM confidence interval.     

Mapping the quantitative trait loci (QTL) for body shape and conformation measurements on BTA1 in Japanese Black cattle

The detection and mapping of segregating quantitative trait loci (QTL) that influence withers height, hip height, hip width, body length, chest width, chest depth, shoulder width, lumbar width, thurl width, pin bone width, rump length, cannon circumference, chest girth, abdominal width and abdominal girth at weaning was conducted on chromosomal regions of bovine chromosome one. The QTL analysis was performed by genotyping half‐sib progeny of five Japanese Black sires using microsatellite DNA markers. Probability coefficients of inheriting allele 1 or 2 from the sire at specific chromosomal locations were computed. The phenotypic data of progeny were regressed on these probability coefficients in a within‐common‐parent regression analysis using a linear model that included fixed effects of sex, parity and season of birth, as well as age as a covariate. F‐statistics were calculated every 1 cM on a linkage map. Permutation tests of 10 000 iterations were conducted to obtain chromosome‐wide significance thresholds. A significant QTL for chest width was detected at 91 cM in family 3. The detection of this QTL boosts the prospects of implementing marker‐assisted selection for body conformation traits in Japanese Black beef cattle.