Journal of Animal Breeding and Genetics最新文献

Prediction error variance (PEV) and prediction accuracy (PA) of breeding values (BVs) are essential for formulating breeding plans and predicting response to selection. However, restricted best linear unbiased prediction method (RBLUP method) carries many unknowns: in particular, the formulas for calculating PEV and PA are not clear. New findings were obtained using the RBLUP method. The uniqueness of RBLUP of BVs was proven. The formulas of PEV and PA for the RBLUP of BVs were derived from restricted mixed model equations. A method was also devised for easily calculating the PEV and PA for the RBLUP of BVs. Finally, the relationship between the RBLUP and ordinary BLUP of BVs was derived. It has become easier to calculate the PEV and PA for the RBLUP of BVs. This method is particularly effective for calculating the PEV and PA when applying the RBLUP method to achieve relative desired changes in all traits. This has also made it possible to predict the response to selection using the RBLUP method.

Understanding the genetic basis of growth and metabolic traits in sheep is crucial for improving production efficiency and sustainability. The current study aimed to estimate the genetic influences, both direct and maternal, on growth rate and Kleiber's ratio traits in Harnali sheep using pedigree data under Bayesian inference. The data pertained to 2404 animals spanned over 24 years (1998-2021). Fixed factors such as birth period, lamb sex and dam's weight at lambing were considered. The traits studied included average daily gains (ADGs) categorised into ADG1 (birth to weaning age), ADG2 (weaning to 6 months of age) and ADG3 (6-12 months of age), as well as corresponding Kleiber's ratios (KR1, KR2 and KR3). Six single-trait animal models were employed to estimate covariance components and heritabilities, integrating direct additive and maternal effects alongside significant fixed factors using THRGIBBS1F90 and POSTGIBBSF90 programmes. Direct heritability estimates were obtained for ADG1 (0.11 ± 0.05), ADG2 (0.06 ± 0.03), ADG3 (0.03 ± 0.03), KR1 (0.07 ± 0.03), KR2 (0.06 ± 0.03) and KR3 (0.05 ± 0.03). Maternal genetic effects have contributed significant particularly to pre-weaning traits. The study identified an antagonistic relationship between direct additive and maternal genetic effects. Positive genetic and phenotypic correlations emphasised the intricate relationship between growth and metabolic efficiency in Harnali sheep. The current study offers critical insights into the genetic basis of growth and metabolic traits in Harnali sheep, ultimately contributing to more efficient and sustainable sheep production systems.

Mastitis in cattle is often caused by microorganism infections in the udder. The three most common pathogens are esculin-positive streptococci (SC+), coagulase-negative staphylococci (CNS), and Escherichia coli (E. coli). In a previous study, 10 SNPs were associated with somatic cell score and mastitis in diverse Holstein populations. We tested these SNPs for their effects on individual pathogen presence. Milk and pathogen samples of 3076 Holstein cows were collected from four farms. Samples were excluded if multiple pathogens were present at the same time. Records of the same pathogen within 14 days of each other were counted as one infection. This resulted in 1129 pathogen-positive samples. Cases and controls were in ratios of 20:80 for SC+, 8:92 for CNS, and 11:89 for E. coli. The lasso, backward, and forward methods were used to narrow down SNPs associated with pathogen presence. The suitability of the SNPs to separate the samples into cases or controls for each pathogen was indicated using ROC curves. The Cochran-Armitage (CAT) and the Jonckheere-Terpstra (JTT) tests evaluated the influence of the SNPs on pathogen presence. Finally, a generalised linear mixed model (GLMM) including fixed environmental effects and a random sire effect was fitted to the binary trait of pathogen presence to test for association. In total, six out of the 10 investigated SNPs showed associations with pathogen presence based on the forward method: Two SNPs each for SC+ (rs41588957, rs41257403) and CNS (rs109934030, rs109441194), and three for E. coli (rs109934030, rs41634110, rs41636878). The CAT and GTT tests linked four SNPs (rs41588957, rs41634110, rs109441194, rs41636878) to pathogen presence, two of which were confirmed with the GLMM (rs41634110, rs109441194), with effects on CNS and E. coli. The SNPs linked to CNS and those linked to E. coli explained 13.2% and 13.8% of the variance, compared to 19% and 18.4%, respectively, of the full model with all 10 SNPs. Half of the SNP genotypes previously linked to lower SCS also decreased the probability for pathogen presence and might therefore be targets not just for lower SCS but for a better pathogen resistance. Trial Registration: Not applicable, no new data were collected for this study.

Genomic selection (GS) and genome-wide association studies (GWAS) have not been investigated in Vietnamese dairy cattle, even for basic milk production traits, largely due to the scarcity of individual phenotype recording in smallholder dairy farms (SDFs). This study aimed to estimate heritability (h²) and test the applicability of GS and GWAS for milk production, body conformation and novel heat tolerance traits using single test day phenotypic data. Thirty-two SDFs located in either the north (a lowland vs. a highland) or the south (a lowland vs. a highland) of Vietnam were each visited for an afternoon and the next morning to collect phenotype data of all lactating cows (n = 345). Tail hair from each cow was sampled for subsequent genotyping with a 50K SNP chip at that same visit. Milk production traits (single-test day) were milk yield (MILK, kg/cow/day), energy corrected milk yield adjusted for body weight (ECMbw, kg/100 kg BW/day), fat (mFA, %), protein (mPR, %) and dry matter (mDM, %). Conformation traits were body weight (BW, kg) and body condition score (BCS, 1 = thin to 5 = obese). Heat tolerance traits were panting score (PS, 0 = normal to 4.5 = extremely heat-stressed) and infrared temperatures (IRTs, °C) at 11 areas on the external body surface of the cow (inner vulval lip, outer vulval surface, inner tail base surface, ocular area, muzzle, armpit area, paralumbar fossa area, fore udder, rear udder, forehoof and hind hoof), assessed by an Infrared Camera. Univariate linear mixed models and a 10-fold cross-validation approach were applied for GS. Univariate single SNP mixed linear models were applied for the GWAS. Estimated h² (using the genotype information to build relationships among animals) were moderate (0.20-0.37) for ECMbw, mFA, mPR, mRE, BW, BCS and IRT at rear udder; low (0.08-0.19) for PS and other IRTs; and very low (≤ 0.07) for MILK, ECM and mDM. Accuracy of genomic estimated breeding values (GEBVs) was low (≤ 0.12) for MILK, ECM, mDM and IRT at hind hoof; and moderate to high (0.32-0.46) for all other traits. The most significant regions on chromosomes (BTA) associated with milk production traits were 0.47-1.18 Mb on BTA14. Moderate to high h² and moderate accuracies of GEBVs for mFA, mPR, ECMbw, BCS, BW, PS and IRTs at rear udder and outer vulval surface suggested that GS using single test day phenotypic data could be applied for these traits. However, a greater sample size is required to decrease the bias of GEBVs by GS and increase the power of detecting significant quantitative trait loci (QTLs) by GWAS.

Recent years have witnessed a remarkable global surge in fish production, with Nile tilapia (Oreochromis niloticus) emerging as a prominent contributor owing to its high demand as a nutritious food source. However, unlike terrestrial species, maintaining genealogical control and collecting phenotypic data in fish farming poses significant challenges, necessitating advancements to support genetic improvement programmes. While conventional methods, such as body measurements using rulers and photographs are prevalent in data collection, the potential of ultrasound-a less invasive and efficient tool for fish measurement-remains underexplored. This study assesses the viability of ultrasonography for genetically selecting carcass characteristics in Nile tilapia. The investigation encompasses data from 897 animals representing 53 full-sib tilapia families maintained in the genetic improvement programme at the Federal University of Lavras. To measure carcass traits, the animals were sedated with benzocaine and ultrasound images were obtained at three distinct points. Subsequently, the animals were euthanised through medullary sectioning for further carcass processing. After evisceration, filleting and skinning, all weights were meticulously recorded. (Co)variance components and genetic parameters of the measured traits were estimated using the Bayesian approach by Gibbs sampling implemented in MTGSAM (Multiple Trait Gibbs Sampling in Animal Models) software. Heritabilities estimated for the studied carcass traits were moderate, ranging from 0.23 to 0.33. Notably, phenotypes derived from ultrasound images demonstrated substantial genetic correlations with fillet yield (0.83-0.92). In conclusion, this study confirms that indirect selection based on ultrasound images is effective and holds promise for integration into tilapia breeding programmes aimed at enhancing carcass yield.

Electronic feeders record feeding behaviour as feed events by tracking the animal's in-out visits to the feeder. Another way to measure feeding behaviour is based on meals. However, the two approaches provide different outcomes. The objectives of this study were to estimate genetic parameters (heritabilities and genetic and phenotypic correlations) for feed event and meal traits, and their genetic and phenotypic correlations with feed efficiency traits in Nellore cattle. The present study analysed six feed event traits (DMI_FE: dry matter intake per feed event, FED: feed event duration, TB_FE: time between feed events, FT_d: feeding time per day, FE_d: feed events per day, and FR: feeding rate), six meal traits (DMI_ME: DMI per meal, MED: meal duration, TB_ME: time between meals, MC: meal criterion, MT_d: meal time per day, and ME_d: meals per day), and three feed efficiency traits (ADG: average daily gain, DMI, and RFI: residual feed intake). The traits were measured in feed efficiency tests of Nellore cattle (age = 280 ± 41 days and body weight = 258 ± 47 kg at enrolment). The MC was calculated for each animal and ranged from 1.70 to 64.0 min, i.e., any pair of feed events separated by less than the MC value was considered part of the same meal. The heritabilities and correlations were estimated by fitting univariate and bivariate animal models, respectively, using single-step genomic BLUP. The highest heritabilities for feed event traits were 0.35 ± 0.06 (FED), 0.39 ± 0.06 (FT_d), and 0.50 ± 0.05 (FT_d), and for meal traits were 0.31 ± 0.06 (MED) and 0.45 ± 0.06 (MT_d). The genetic correlation between feed event traits and meal traits were weak. FR, FED, and FT_d had moderate genetic correlations with RFI (-0.56 ± 0.11, 0.44 ± 0.11, 0.60 ± 0.08, respectively). These results indicate that more efficient animals spent less time at the feeder per feed event and per day, and eat faster compared to less efficient animals. In conclusion, feed event and meal traits must be treated as distinct groups of traits since the genetic and phenotypic correlations were, in general, weak to moderate. Among feed event versus meal traits, feed event traits are more favourable to explain the genetic relationships of feeding behaviour with feed efficiency-related traits.

Aneuploidy, a genetic condition characterised by the deletion (monosomy) or duplication (trisomy) of a chromosome, has been extensively studied in humans, particularly in the context of trisomy on chromosome 21, also known as Down syndrome. Research on autosomal aneuploidy in live-born cattle has been limited to case reports, resulting in a lack of prevalence estimates of aneuploidy in cattle. Furthermore, the viability or lethality of aneuploidy on specific autosomes in cattle has not been well documented. The objective of this study was to estimate the prevalence of autosomal aneuploidy in a large population of new-born and juvenile beef and dairy cattle using single nucleotide polymorphism (SNP) chip genotype intensity data. Of the population of 779,138 cattle genotyped when younger than 15 months of age, 139 cattle (i.e., 0.017%) were diagnosed with one case of autosomal trisomy. Trisomy in only 10 different autosomes were detected (BTA 4, 6, 12, 15, 20, 24, 26, 27, 28 and 29) albeit the one case of trisomy detected on Bos taurus autosome (BTA) 4 was in an additional population of 341,927 cattle that were genotyped at > 15 months of age and was therefore excluded from prevalence estimates to minimise bias. The prevalence of trisomy per chromosome was generally inversely related to the length of the chromosome. Although the number of affected individuals was few, there was no evidence of differences in prevalence by breed, inbreeding level or parental age. The parental origin of the detected cases of trisomy was maternal for 92% of the cases. No cases of monosomy were detected despite the large dataset, which included calves genotyped at birth, indicating the potential lethal nature of monosomy in cattle. Cytogenetic testing was used to verify three of the animals with detected autosomal trisomy who were still alive. Eighteen of the 139 animals identified with autosomal trisomy were recorded as being stillborn, resulting in a prevalence of autosomal aneuploidy in live-born cattle of 0.015%. Of the 121 live-born cattle with autosomal trisomy, a total of 68 died on farm at, on average (standard deviation), 6.8 (8.7) months of age. All animals with autosomal trisomy on BTA 6, 12, 15, 20 or 24 were either stillborn or died on farm within 15 days of birth. This study is the first report of trisomy on BTA 4, 6, 15, 20 and 27 in live-born cattle, as well as the first to document fertile cows with trisomy on BTA 4, 27 or 28. Given that genotype intensity SNP data from SNP-chips are readily available, identifying animals affected with autosomal aneuploidy as well as quantifying and monitoring the incidence can be easily undertaken.

This study aimed to estimate variance components (VCs) for growth and reproductive traits in Nellore cattle using two relationship matrices (pedigree relationship A matrix and pedigree plus genomic relationship H matrix), and records collected before and after genomic selection (GS) implementation. The study also evaluated how genomic breeding values (GEBV) are affected by variance components and discarding old records. The analysed traits were weight at 120 days (W120), weight and scrotal circumference at 450 days (W450 and SC450, respectively). Three datasets were used to estimate VCs, including all phenotypic information (All) or records for animals born before or after GS implementation (Before or After datasets, respectively). Both relationship matrices were considered for VC estimation, the A matrix was used in all three datasets and VC from each combination were named as A_Before, A_After, and A_All). The H was used in two datasets: H_All and H_After. Different VCs were used for GEBV prediction through ssGBLUP. This step used two possible Datasets, using all available phenotypic data (Dataset 1) or just records collected since GS implementation (Dataset 2). Validation was conducted using accuracy, bias and dispersion according to the LR method and prediction accuracy from corrected phenotypes. The heritability of all traits increased from A_Before to A_After, while estimates for A_All were intermediary. In the previous order, the estimates were 0.16, 0.17, and 0.15 for W120; 0.31, 0.39, and 0.35 for W450; 0.35, 0.47, and 0.41 for SC. For W450 and SC, using the H matrix reduced the heritability (0.33 and 0.32 for W450; 0.41 and 0.38 for SC for H_After and H_All, respectively). For W120, Dataset1 and VCs from A_After showed the highest accuracy for direct and maternal GEBV (0.953 and 0.868). For W450, Dataset 1 and VC from H_After allowed the highest accuracy (0.854) but use Dataset 2 and same VC source yield similar value (0.846). For SC, Dataset 2 with VC from H_After showed the highest accuracy (0.925). To use Dataset 2 does not cause important changes in bias or dispersion with respect to Dataset 1. The VC and genetic parameters changed for W120, W450, and SC450, using records before or after the GS implementation. For W450 and SC450, genetic variance and heritability estimates increased with the use of GS. For W120, genomic predictions were more accurate using A for VC estimation. Accuracy gains were observed for W450 and SC450 using H in VC estimation and/or discarding records before GS. It is possible to discard phenotypic records before GS implementation without generating bias or dispersion in the GEBV of young candidates.

Genomic selection is widely implemented in livestock breeding programmes across species. Its potential is also evident for sheep breeding; however, it has several limitations, particularly because of the high genetic diversity across and within sheep breeds. In Germany, the predominant sheep breed is the Merino sheep. Until now, there has been no use of genomic selection in the German Merino sheep breeding programme. In this simulation study, different genomic selection strategies were compared with a reference scenario with a breeding value estimation based on pedigree BLUP. A simplified version of the German Merino sheep breeding programme, including a health and a production trait in the breeding goal, was simulated via the R package Modular Breeding Program Simulator (MoBPS). Real genotype data were used to create a population specific simulation. The reference scenario was compared with several alternative scenarios in which selection was based on single-step GBLUP (ssGBLUP) breeding value estimation with varying genotyping strategies. In addition to scenarios in which all male and all male plus all female lambs were genotyped, scenarios with a preselection of lambs, that is only a certain proportion (top 25%, top 50%) genotyped, were simulated. The results revealed that genetic gain increased with increasing numbers of available genotypes. However, marginal gains decreased with increasing numbers of genotypes. Compared with the reference scenario, genotyping the top 25% of male lambs increased the genetic gain for the breeding ram population by 13% for both traits whereas genotyping the top 50% of male lambs or all male lambs led to increases of 18% (17%) or 26% (21%) for the health (production) trait, respectively. The potential of genotyping females in addition to male lambs was less evident on the male side with no significant differences between the scenarios with different proportions of genotyped females. The results have shown that genomic selection can be a valuable tool to increase genetic gain in the German Merino sheep population and that the genotyping of a certain proportion of animals might lead to substantial improvement over pedigree-based breeding value estimation. Nevertheless, further studies, especially economic evaluations, are needed before practical implementation.