Machine learning methods for genomic prediction of cow behavioral traits measured by automatic milking systems in North American Holstein cattle.

Identifying genome-enabled methods that provide more accurate genomic prediction is crucial when evaluating complex traits such as dairy cow behavior. In this study, we aimed to compare the predictive performance of traditional genomic prediction methods and deep learning algorithms for genomic prediction of milking refusals (MREF) and milking failures (MFAIL) in North American Holstein cows measured by automatic milking systems (milking robots). A total of 1,993,509 daily records from 4,511 genotyped Holstein cows were collected by 36 milking robot stations. After quality control, 57,600 SNPs were available for the analyses. Four genomic prediction methods were considered: Bayesian least absolute shrinkage and selection operator (LASSO), multiple layer perceptron (MLP), convolutional neural network (CNN), and GBLUP. We implemented the first 3 methods using the Keras and TensorFlow libraries in Python (v.3.9) but the GBLUP method was implemented using the BLUPF90+ family programs. The accuracy of genomic prediction (mean square error) for MREF and MFAIL was 0.34 (0.08) and 0.27 (0.08) based on LASSO, 0.36 (0.09) and 0.32 (0.09) for MLP, 0.37 (0.08) and 0.30 (0.09) for CNN, and 0.35 (0.09) and 0.31(0.09) based on GBLUP, respectively. Additionally, we observed a lower reranking of top selected individuals based on the MLP versus CNN methods compared with the other approaches for both MREF and MFAIL. Although the deep learning methods showed slightly higher accuracies than GBLUP, the results may not be sufficient to justify their use over traditional methods due to their higher computational demand and the difficulty of performing genomic prediction for nongenotyped individuals using deep learning procedures. Overall, this study provides insights into the potential feasibility of using deep learning methods to enhance genomic prediction accuracy for behavioral traits in livestock. Further research is needed to determine their practical applicability to large dairy cattle breeding programs.

Investigating relationships between the host genome, rumen microbiome, and dairy cow feed efficiency using mediation analysis with structural equation modeling.

The rumen microbiome is crucial for converting feed into absorbable nutrients used for milk synthesis, and the efficiency of this process directly impacts the profitability and sustainability of the dairy industry. Recent studies have found that the rumen microbial composition explains part of the variation in feed efficiency traits, including dry matter intake, milk energy, and residual feed intake. The main goal of this study was to reveal relationships between the host genome, rumen microbiome, and dairy cow feed efficiency using structural equation models. Our specific objectives were to (i) infer the mediation effects of the rumen microbiome on feed efficiency traits, (ii) estimate the direct and total heritability of feed efficiency traits, and (iii) calculate the direct and total breeding values of feed efficiency traits. Data consisted of dry matter intake, milk energy, and residual feed intake records, SNP genotype data, and 16S rRNA rumen microbial abundances from 448 mid-lactation Holstein cows from 2 research farms. We implemented structural equation models such that the host genome directly affects the phenotype (GP → P) and the rumen microbiome (GM → P), while the microbiome affects the phenotype (M → P), partially mediating the effect of the host genome on the phenotype (G → M → P). We found that 7 to 30% of microbes within the rumen microbial community had structural coefficients different from zero. We classified these microbes into 3 groups that could have different uses in dairy farming. Microbes with heritability <0.10 but significant causal effects on feed efficiency are attractive for external interventions. On the other hand, 2 groups of microbes with heritability ≥0.10, significant causal effects, and genetic covariances and causal effects with the same or opposite sign to feed efficiency are attractive for selective breeding, improving or decreasing the trait heritability and response to selection, respectively. In general, the inclusion of the different microbes in genomic models tends to decrease the trait heritability rather than increase it, ranging from -15% to +5%, depending on the microbial group and phenotypic trait. Our findings provide more understanding to target rumen microbes that can be manipulated, either through selection or management interventions, to improve feed efficiency traits.

Longitudinal analysis of bovine mammary gland development.

Many studies on bovine mammary glands focus on one stage of development. Often missing in those studies are repeated measures of development from the same animals. As milk production is directly affected by amount of parenchymal tissue within the udder, understanding mammary gland growth along with visualization of its structures during development is essential. Therefore, analysis of ultrasound and histology data from the same animals would result in better understanding of mammary development over time. Thus, this research aimed to describe mammary gland development using non-invasive and invasive tools to delineate growth rate of glandular tissue responsible for potential future milk production. Mammary gland ultrasound images, biopsy samples, and blood samples were collected from 36 heifer dairy calves beginning at 10 weeks of age, and evaluated at 26, 39, and 52 weeks. Parenchyma was quantified at 10 weeks of age using ultrasound imaging and histological evaluation, and average echogenicity was utilized to quantify parenchyma at later stages of development. A significant negative correlation was detected between average echogenicity of parenchyma at 10 weeks and total adipose as a percent of histological whole tissue at 52 weeks. Additionally, a negative correlation between average daily gain at 10 and 26 weeks and maximum echogenicity at 52 weeks was present. These results suggest average daily gain and mammary gland development prior to 39 weeks of age is associated with development of the mammary gland after 39 weeks. These findings could be predictors of future milk production, however this must be further explored.

Genomic prediction for meat and carcass traits in Nellore cattle using a Markov blanket algorithm.

This study was carried out to evaluate the advantage of preselecting SNP markers using Markov blanket algorithm regarding the accuracy of genomic prediction for carcass and meat quality traits in Nellore cattle. This study considered 3675, 3680, 3660 and 524 records of rib eye area (REA), back fat thickness (BF), rump fat (RF), and Warner-Bratzler shear force (WBSF), respectively, from the Nellore Brazil Breeding Program. The animals have been genotyped using low-density SNP panel (30 k), and subsequently imputed for arrays with 777 k SNPs. Four Bayesian specifications of genomic regression models, namely Bayes A, Bayes B, Bayes Cπ and Bayesian Ridge Regression methods were compared in terms of prediction accuracy using a five folds cross-validation. Prediction accuracy for REA, BF and RF was all similar using the Bayesian Alphabet models, ranging from 0.75 to 0.95. For WBSF, the predictive ability was higher using Bayes B (0.47) than other methods (0.39 to 0.42). Although the prediction accuracies using Markov blanket of SNP markers were lower than those using all SNPs, for WBSF the relative gain was lower than 13%. With a subset of informative SNPs markers, identified using Markov blanket, probably, is possible to capture a large proportion of the genetic variance for WBSF. The development of low-density and customized arrays using Markov blanket might be cost-effective to perform a genomic selection for this trait, increasing the number of evaluated animals, improving the management decisions based on genomic information and applying genomic selection on a large scale.

Analysis of the causal structure of traits involved in sow lactation feed efficiency.

BACKGROUND: Feed efficiency during lactation involves a set of phenotypic traits that form a complex system, with some traits exerting causal effects on the others. Information regarding such interrelationships can be used to predict the effect of external interventions on the system, and ultimately to optimize management practices and multi-trait selection strategies. Structural equation models can be used to infer the magnitude of the different causes of such interrelationships. The causal network necessary to fit structural equation models can be inferred using the inductive causation (IC) algorithm. By implementing these statistical tools, we inferred the causal association between the main energy sources and sinks involved in sow lactation feed efficiency for the first time, i.e., daily lactation feed intake (dLFI) in kg/day, daily sow weight balance (dSWB) in kg/day, daily litter weight gain (dLWG) in kg/day, daily back fat thickness balance (dBFTB) in mm/day, and sow metabolic body weight (SMBW) in kg0.75. Then, we tested several selection strategies based on selection indices, with or without dLFI records, to improve sow efficiency during lactation. RESULTS: The IC algorithm using 95% highest posterior density (HPD95%) intervals resulted in a fully directed acyclic graph, in which dLFI and dLWG affected dSWB, the posterior mean of the corresponding structural coefficients (PMλ) being 0.12 and - 0.03, respectively. In turn, dSWB influenced dBFTB and SMBW, with PMλ equal to 0.70 and - 1.22, respectively. Multiple indirect effects contributed to the variances and covariances among the analyzed traits, with the most relevant indirect effects being those involved in the association between dSWB and dBFTB and between dSWB and SMBW. Selection strategies with or without phenotypic information on dLFI, or that hold this trait constant, led to the same pattern and similar responses in dLFI, dSWB, and dLWG. CONCLUSIONS: Selection based on an index including only dBFTB and dLWG records can reduce dLFI, keep dSWB constant or increase it, and increase dLWG. However, a favorable response for all three traits is probably not achievable. Holding the amount of feed provided to the sows constant did not offer an advantage in terms of response over the other strategies.

Economic values of reproductive, growth, feed efficiency and carcass traits in Nellore cattle.

A bioeconomic model was developed to calculate the economic value (ev) of reproductive and growth performance, feed efficiency and carcass traits of a seedstock Nellore herd. Data from a full-cycle cattle operation (1,436 dams) located in the Brazilian Cerrado were assessed. The ev was calculated by the difference in profit before and after one-unit improvement in the trait, with others remaining unchanged. The ev was standardized by the phenotypic standard deviation of each trait. Preweaning average daily gain (ADG) was the most economically important trait evaluated (R$ 58.04/animal/year), followed by age at first calving (R$ 44.35), postweaning ADG (R$ 31.43), weight at 450 days (R$ 25.36), accumulated productivity (R$ 21.43), ribeye area (R$ 21.35), calving interval (R$ 19.97), feed efficiency (R$ 15.24), carcass dressing per cent (R$ 8.27), weight at 120 days (R$ 6.22), weight at 365 days (R$ 6.06), weight at weaning (210 days, R$ 5.82), stayability (R$ 5.70) and the probability of early calving (R$ 0.32). The effects of all traits on profits are evidence that their selection may result in the economic and genetic progress of the herd if there is genetic variability.

Predictive assessment of single-step BLUP with linear and non-linear similarity RKHS kernels: A case study in chickens.

Single-step GBLUP (ssGBLUP) to obtain genomic prediction was proposed in 2009. Many studies have investigated ssGBLUP in genomic selection in animals and plants using a standard linear kernel (similarity matrix) called genomic relationship matrix (G). More general kernels should allow capturing non-additive effects as well, whereas GBLUP is based on additive gene action. In this study, we generalized ssBLUP to accommodate two non-linear kernels, the averaged Gaussian kernel (AK) and the recently developed arc-cosine deep kernel (DK). We evaluated the methodology using body weight (BW) and hen-housing production (HHP) traits, recorded on a sample of phenotyped and genotyped commercial broiler chickens. There were, thus, different ssGBLUP models corresponding to G, AK and DK. We used random replication of training (TRN) and testing (TST) layouts at different genotyping rates (20%, 40%, 60% and 80% of all birds) in three selective genotyping scenarios. The selections were genotyping the youngest individuals in the pedigree (YS), random genotyping (RS) and genotyping based on parent average (PA). Predictive abilities were measured using rank correlations between the observed and the predictive phenotypic values in TST for each random partition. Prediction accuracy was influenced by the type of kernel when a large proportion of birds was genotyped. An advantage of non-linear kernels (AK and DK) was more apparent when 60 and 80% of birds had been genotyped. For BW, the lowest rank correlations were obtained with G (0.093 ± 0.015 using RS by 20% genotyped individuals) and the highest values with DK (0.320 ± 0.016 in the PA setting with 80% genotyped individuals). For HHP, the lowest and highest rank correlations were obtained by AK with 20% and 80% genotyped individuals, 0.071 ± 0.016 (in RS) and 0.23 ± 0.016 (in PA) respectively. Our results indicated that AK and DK are more effective than G when a large proportion of the target population is genotyped. Our expectation is that ssGBLUP with AK or DK models can perform even better than G when non-additive genetic effects influence the underlying variability of complex traits.

Enviromics in breeding: applications and perspectives on envirotypic-assisted selection.

KEY MESSAGE: We propose the application of enviromics to breeding practice, by which the similarity among sites assessed on an "omics" scale of environmental attributes drives the prediction of unobserved genotype performances. Genotype by environment interaction (GEI) studies in plant breeding have focused mainly on estimating genetic parameters over a limited number of experimental trials. However, recent geographic information system (GIS) techniques have opened new frontiers for better understanding and dealing with GEI. These advances allow increasing selection accuracy across all sites of interest, including those where experimental trials have not yet been deployed. Here, we introduce the term enviromics, within an envirotypic-assisted breeding framework. In summary, likewise genotypes at DNA markers, any particular site is characterized by a set of "envirotypes" at multiple "enviromic" markers corresponding to environmental variables that may interact with the genetic background, thus providing informative breeding re-rankings for optimized decisions over different environments. Based on simulated data, we illustrate an index-based enviromics method (the "GIS-GEI") which, due to its higher granular resolution than standard methods, allows for: (1) accurate matching of sites to their most appropriate genotypes; (2) better definition of breeding areas that have high genetic correlation to ensure selection gains across environments; and (3) efficient determination of the best sites to carry out experiments for further analyses. Environmental scenarios can also be optimized for productivity improvement and genetic resources management, especially in the current outlook of dynamic climate change. Envirotyping provides a new class of markers for genetic studies, which are fairly inexpensive, increasingly available and transferable across species. We envision a promising future for the integration of enviromics approaches into plant breeding when coupled with next-generation genotyping/phenotyping and powerful statistical modeling of genetic diversity.

Nonlinear modeling to describe the pattern of 15 milk protein and nonprotein compounds over lactation in dairy cows.

The protein profile of milk includes several caseins, whey proteins, and nonprotein nitrogen compounds, which influence milk's value for human nutrition and its cheesemaking properties for the dairy industry. To fill in the gap in current knowledge of the patterns of these individual nitrogenous compounds throughout lactation, we tested the ability of a parametric nonlinear lactation model to describe the pattern of each N compound expressed qualitatively (as % of total milk N), quantitatively (in g/L milk), and as daily yield (in g/d). The lactation model was tested on a data set of detailed milk nitrogenous compound profiles (15 fractions-12 protein traits and 3 nonproteins-for each expression mode: 45 traits) obtained from 1,342 cows reared in 41 multibreed herds. Our model was a modified version of Wilmink's model, often used for describing milk yield during lactation because of its reliability and ease of parameter interpretation from a biological point of view. We allowed the sign of the persistency coefficient (parameter c) that explained the variation in the long-term milk component (parameter a) to be positive or negative. We also allowed the short-term milk component (parameter b) to be positive or negative, and we estimated a specific speed of adaptation parameter (parameter k) for each trait rather than assumed a value a priori, as in the original model (k = 0.05). These 4 parameters were included in a nonlinear mixed model with cow breed and parity order as fixed effects, and herd-date as random. Combinations of the positive and negative signs of the b and c parameters allowed us to identify 4 differently shaped lactation curves, all found among the patterns exhibited by the nitrogenous fractions as follows: the "zenith" curve (with a maximum peak; for milk yield and 10 other N traits), the "nadir" curve (with a minimum point; for 20 traits, including almost all those expressed in g/L of milk), the "downward" curve (continuously decreasing; for 14 traits, including almost all those in g/d), and the "upward" curve (continuously increasing; only for κ-casein, in % N). Direct estimation of the k parameters specific to each trait showed the large variability in the adaptation speed of fresh cows and greatly increased the model's flexibility. The results indicated that nonlinear parametric mathematical models can effectively describe the different and complex patterns exhibited by individual nitrogenous fractions during lactation; therefore, they could be useful tools for interpreting milk composition variations during lactation.

Structural equation modeling for unraveling the multivariate genomic architecture of milk proteins in dairy cattle.

The aims of this study were to investigate potential functional relationships among milk protein fractions in dairy cattle and to carry out a structural equation model (SEM) GWAS to provide a decomposition of total SNP effects into direct effects and effects mediated by traits that are upstream in a phenotypic network. To achieve these aims, we first fitted a mixed Bayesian multitrait genomic model to infer the genomic correlations among 6 milk nitrogen fractions [4 caseins (CN), namely κ-, ß-, αS1-, and αS2-CN, and 2 whey proteins, namely ß-lactoglobulin (ß-LG) and α-lactalbumin (α-LA)], in a population of 989 Italian Brown Swiss cows. Animals were genotyped with the Illumina BovineSNP50 Bead Chip v.2 (Illumina Inc.). A Bayesian network approach using the max-min hill-climbing (MMHC) algorithm was implemented to model the dependencies or independence among traits. Strong and negative genomic correlations were found between ß-CN and αS1-CN (-0.706) and between ß-CN and κ-CN (-0.735). The application of the MMHC algorithm revealed that κ-CN and ß-CN seemed to directly or indirectly influence all other milk protein fractions. By integrating multitrait model GWAS and SEM-GWAS, we identified a total of 127 significant SNP for κ-CN, 89 SNP for ß-CN, 30 SNP for αS1-CN, and 14 SNP for αS2-CN (mostly shared among CN and located on Bos taurus autosome 6) and 15 SNP for ß-LG (mostly located on Bos taurus autosome 11), whereas no SNP passed the significance threshold for α-LA. For the significant SNP, we assessed and quantified the contribution of direct and indirect paths to total marker effect. Pathway analyses confirmed that common regulatory mechanisms (e.g., energy metabolism and hormonal and neural signals) are involved in the control of milk protein synthesis and metabolism. The information acquired might be leveraged for setting up optimal management and selection strategies aimed at improving milk quality and technological characteristics in dairy cattle.

Would large dataset sample size unveil the potential of deep neural networks for improved genome-enabled prediction of complex traits? The case for body weight in broilers.

BACKGROUND: Deep neural networks (DNN) are a particular case of artificial neural networks (ANN) composed by multiple hidden layers, and have recently gained attention in genome-enabled prediction of complex traits. Yet, few studies in genome-enabled prediction have assessed the performance of DNN compared to traditional regression models. Strikingly, no clear superiority of DNN has been reported so far, and results seem highly dependent on the species and traits of application. Nevertheless, the relatively small datasets used in previous studies, most with fewer than 5000 observations may have precluded the full potential of DNN. Therefore, the objective of this study was to investigate the impact of the dataset sample size on the performance of DNN compared to Bayesian regression models for genome-enable prediction of body weight in broilers by sub-sampling 63,526 observations of the training set. RESULTS: Predictive performance of DNN improved as sample size increased, reaching a plateau at about 0.32 of prediction correlation when 60% of the entire training set size was used (i.e., 39,510 observations). Interestingly, DNN showed superior prediction correlation using up to 3% of training set, but poorer prediction correlation after that compared to Bayesian Ridge Regression (BRR) and Bayes Cπ. Regardless of the amount of data used to train the predictive machines, DNN displayed the lowest mean square error of prediction compared to all other approaches. The predictive bias was lower for DNN compared to Bayesian models, across all dataset sizes, with estimates close to one with larger sample sizes. CONCLUSIONS: DNN had worse prediction correlation compared to BRR and Bayes Cπ, but improved mean square error of prediction and bias relative to both Bayesian models for genome-enabled prediction of body weight in broilers. Such findings, highlights advantages and disadvantages between predictive approaches depending on the criterion used for comparison. Furthermore, the inclusion of more data per se is not a guarantee for the DNN to outperform the Bayesian regression methods commonly used for genome-enabled prediction. Nonetheless, further analysis is necessary to detect scenarios where DNN can clearly outperform Bayesian benchmark models.

Integration of single nucleotide variants and whole-genome DNA methylation profiles for classification of rheumatoid arthritis cases from controls.

This study evaluated the use of multiomics data for classification accuracy of rheumatoid arthritis (RA). Three approaches were used and compared in terms of prediction accuracy: (1) whole-genome prediction (WGP) using SNP marker information only, (2) whole-methylome prediction (WMP) using methylation profiles only, and (3) whole-genome/methylome prediction (WGMP) with combining both omics layers. The number of SNP and of methylation sites varied in each scenario, with either 1, 10, or 50% of these preselected based on four approaches: randomly, evenly spaced, lowest p value (genome-wide association or epigenome-wide association study), and estimated effect size using a Bayesian ridge regression (BRR) model. To remove effects of high levels of pairwise linkage disequilibrium (LD), SNPs were also preselected with an LD-pruning method. Five Bayesian regression models were studied for classification, including BRR, Bayes-A, Bayes-B, Bayes-C, and the Bayesian LASSO. Adjusting methylation profiles for cellular heterogeneity within whole blood samples had a detrimental effect on the classification ability of the models. Overall, WGMP using Bayes-B model has the best performance. In particular, selecting SNPs based on LD-pruning with 1% of the methylation sites selected based on BRR included in the model, and fitting the most significant SNP as a fixed effect was the best method for predicting disease risk with a classification accuracy of 0.975. Our results showed that multiomics data can be used to effectively predict the risk of RA and identify cases in early stages to prevent or alter disease progression via appropriate interventions.

Improving genomic prediction accuracy for meat tenderness in Nellore cattle using artificial neural networks.

The goal of this study was to compare the predictive performance of artificial neural networks (ANNs) with Bayesian ridge regression, Bayesian Lasso, Bayes A, Bayes B and Bayes Cπ in estimating genomic breeding values for meat tenderness in Nellore cattle. The animals were genotyped with the Illumina Bovine HD Bead Chip (HD, 777K from 90 samples) and the GeneSeek Genomic Profiler (GGP Indicus HD, 77K from 485 samples). The quality control for the genotypes was applied on each Chip and comprised removal of SNPs located on non-autosomal chromosomes, with minor allele frequency <5%, deviation from HWE (p < 10-6 ), and with linkage disequilibrium >0.8. The FImpute program was used for genotype imputation. Pedigree-based analyses indicated that meat tenderness is moderately heritable (0.35), indicating that it can be improved by direct selection. Prediction accuracies were very similar across the Bayesian regression models, ranging from 0.20 (Bayes A) to 0.22 (Bayes B) and 0.14 (Bayes Cπ) to 0.19 (Bayes A) for the additive and dominance effects, respectively. ANN achieved the highest accuracy (0.33) of genomic prediction of genetic merit. Even though deep neural networks are recognized to deliver more accurate predictions, in our study ANN with one single hidden layer, 105 neurons and rectified linear unit (ReLU) activation function was sufficient to increase the prediction of genetic merit for meat tenderness. These results indicate that an ANN with relatively simple architecture can provide superior genomic predictions for meat tenderness in Nellore cattle.

Causal phenotypic networks for egg traits in an F2 chicken population.

Traditional single-trait genetic analyses, such as quantitative trait locus (QTL) and genome-wide association studies (GWAS), have been used to understand genotype-phenotype relationships for egg traits in chickens. Even though these techniques can detect potential genes of major effect, they cannot reveal cryptic causal relationships among QTLs and phenotypes. Thus, to better understand the relationships involving multiple genes and phenotypes of interest, other data analysis techniques must be used. Here, we utilized a QTL-directed dependency graph (QDG) mapping approach for a joint analysis of chicken egg traits, so that functional relationships and potential causal effects between them could be investigated. The QDG mapping identified a total of 17 QTLs affecting 24 egg traits that formed three independent networks of phenotypic trait groups (eggshell color, egg production, and size and weight of egg components), clearly distinguishing direct and indirect effects of QTLs towards correlated traits. For example, the network of size and weight of egg components contained 13 QTLs and 18 traits that are densely connected to each other. This indicates complex relationships between genotype and phenotype involving both direct and indirect effects of QTLs on the studied traits. Most of the QTLs were commonly identified by both the traditional (single-trait) mapping and the QDG approach. The network analysis, however, offers additional insight regarding the source and characterization of pleiotropy affecting egg traits. As such, the QDG analysis provides a substantial step forward, revealing cryptic relationships among QTLs and phenotypes, especially regarding direct and indirect QTL effects as well as potential causal relationships between traits, which can be used, for example, to optimize management practices and breeding strategies for the improvement of the traits.

Applying family analyses to electronic health records to facilitate genetic research.

Motivation: Pedigree analysis is a longstanding and powerful approach to gain insight into the underlying genetic factors in human health, but identifying, recruiting and genotyping families can be difficult, time consuming and costly. Development of high throughput methods to identify families and foster downstream analyses are necessary. Results: This paper describes simple methods that allowed us to identify 173 368 family pedigrees with high probability using basic demographic data available in most electronic health records (EHRs). We further developed and validate a novel statistical method that uses EHR data to identify families more likely to have a major genetic component to their diseases risk. Lastly, we showed that incorporating EHR-linked family data into genetic association testing may provide added power for genetic mapping without additional recruitment or genotyping. The totality of these results suggests that EHR-linked families can enable classical genetic analyses in a high-throughput manner. Availability and implementation: Pseudocode is provided as supplementary information. Contact: HEBBRING.SCOTT@marshfieldresearch.org. Supplementary information: Supplementary data are available at Bioinformatics online.

Multivariate genome-wide association analysis identifies novel and relevant variants associated with anterior cruciate ligament rupture risk in the dog model.

BACKGROUND: Anterior cruciate ligament rupture (ACLR) is a debilitating and potentially life-changing condition in humans, as there is a high prevalence of early-onset osteoarthritis after injury. Identification of high-risk individuals before they become patients is important, as post-treatment lifetime burden of ACLR in the USA ranges from $7.6 to $17.7 billion annually. ACLR is a complex disease with multiple risk factors including genetic predisposition. Naturally occurring ACLR in the dog is an excellent model for human ACLR, as risk factors and disease characteristics in humans and dogs are similar. In a univariate genome-wide association study (GWAS) of 237 Labrador Retrievers, we identified 99 ACLR candidate loci. It is likely that additional variants remain to be identified. Joint analysis of multiple correlated phenotypes is an underutilized technique that increases statistical power, even when only one phenotype is associated with the trait. Proximal tibial morphology has been shown to affect ACLR risk in both humans and dogs. In the present study, tibial plateau angle (TPA) and relative tibial tuberosity width (rTTW) were measured on bilateral radiographs from purebred Labrador Retrievers that were recruited to our initial GWAS. We performed a multivariate genome wide association analysis of ACLR status, TPA, and rTTW. RESULTS: Our analysis identified 3 loci with moderate evidence of association that were not previously associated with ACLR. A locus on Chr1 associated with both ACLR and rTTW is located within ROR2, a gene important for cartilage and bone development. A locus on Chr4 associated with both ACLR and TPA resides within DOCK2, a gene that has been shown to promote immune cell migration and invasion in synovitis, an important predictor of ACLR. A third locus on Chr23 associated with only ACLR is located near a long non-coding RNA (lncRNA). LncRNA's are important for regulation of gene transcription and translation. CONCLUSIONS: These results did not overlap with our previous GWAS, which is reflective of the different methods used, and supports the need for further work. The results of the present study are highly relevant to ACLR pathogenesis, and identify potential drug targets for medical treatment.

Bayesian Variable Selection in Multilevel Item Response Theory Models with Application in Genomics.

The goal of this paper is to present an implementation of stochastic search variable selection (SSVS) to multilevel model from item response theory (IRT). As experimental settings get more complex and models are required to integrate multiple (and sometimes massive) sources of information, a model that can jointly summarize and select the most relevant characteristics can provide better interpretation and a deeper insight into the problem. A multilevel IRT model recently proposed in the literature for modeling multifactorial diseases is extended to perform variable selection in the presence of thousands of covariates using SSVS. We derive conditional distributions required for such a task as well as an acceptance-rejection step that allows for the SSVS in high dimensional settings using a Markov Chain Monte Carlo algorithm. We validate the variable selection procedure through simulation studies, and illustrate its application on a study with genetic markers associated with the metabolic syndrome.

A predictive assessment of genetic correlations between traits in chickens using markers.

BACKGROUND: Genomic selection has been successfully implemented in plant and animal breeding programs to shorten generation intervals and accelerate genetic progress per unit of time. In practice, genomic selection can be used to improve several correlated traits simultaneously via multiple-trait prediction, which exploits correlations between traits. However, few studies have explored multiple-trait genomic selection. Our aim was to infer genetic correlations between three traits measured in broiler chickens by exploring kinship matrices based on a linear combination of measures of pedigree and marker-based relatedness. A predictive assessment was used to gauge genetic correlations. METHODS: A multivariate genomic best linear unbiased prediction model was designed to combine information from pedigree and genome-wide markers in order to assess genetic correlations between three complex traits in chickens, i.e. body weight at 35 days of age (BW), ultrasound area of breast meat (BM) and hen-house egg production (HHP). A dataset with 1351 birds that were genotyped with the 600 K Affymetrix platform was used. A kinship kernel (K) was constructed as K = λ G + (1 - λ)A, where A is the numerator relationship matrix, measuring pedigree-based relatedness, and G is a genomic relationship matrix. The weight (λ) assigned to each source of information varied over the grid λ = (0, 0.2, 0.4, 0.6, 0.8, 1). Maximum likelihood estimates of heritability and genetic correlations were obtained at each λ, and the "optimum" λ was determined using cross-validation. RESULTS: Estimates of genetic correlations were affected by the weight placed on the source of information used to build K. For example, the genetic correlation between BW-HHP and BM-HHP changed markedly when λ varied from 0 (only A used for measuring relatedness) to 1 (only genomic information used). As λ increased, predictive correlations (correlation between observed phenotypes and predicted breeding values) increased and mean-squared predictive error decreased. However, the improvement in predictive ability was not monotonic, with an optimum found at some 0 < λ < 1, i.e., when both sources of information were used together. CONCLUSIONS: Our findings indicate that multiple-trait prediction may benefit from combining pedigree and marker information. Also, it appeared that expected correlated responses to selection computed from standard theory may differ from realized responses. The predictive assessment provided a metric for performance evaluation as well as a means for expressing uncertainty of outcomes of multiple-trait selection.

Causal effect of prolificacy on milk yield in dairy sheep using propensity score.

In animal production, it is often important to investigate causal relationships among variables. The gold standard tool for such investigation is randomized experiments. However, randomized experiments may not always be feasible, possible, or cost effective or reflect real-world farm conditions. Sometimes it is necessary to infer effects from farm-recorded data. Inferring causal effects between variables from field data is challenging because the association between them may arise not only from the effect of one on another but also from confounding background factors. Propensity score (PS) methods address this issue by correcting for confounding in different levels of the causal variable, which allows unbiased inference of causal effects. Here the objective was to estimate the causal effect of prolificacy on milk yield (MY) in dairy sheep using PS based on matched samples. Data consisted of 4,319 records from 1,534 crossbred ewes. Confounders were lactation number (first, second, and third through sixth) and dairy breed composition (<0.5, 0.5-0.75, and >0.75 of East Friesian or Lacaune). The causal variable prolificacy was considered as 2 levels (single or multiple lambs at birth). The outcome MY represented the volume of milk produced in the whole lactation. Pairs of single- and multiple-birth ewes (1,166) with similar PS were formed. The matching process diminished major discrepancies in the distribution of prolificacy for each confounder variable indicating bias reduction (cutoff standardized bias = 20%). The causal effect was estimated as the average difference within pairs. The effect of prolificacy on MY per lactation was 20.52 L of milk with a simple matching estimator and 12.62 L after correcting for remaining biases. A core advantage of causal over probabilistic approaches is that they allow inference of how variables would react as a result of external interventions (e.g., changes in the production system). Therefore, results imply that management and decision-making practices increasing prolificacy would positively affect MY, which is important knowledge at the farm level. Farm-recorded data can be a valuable source of information given its low cost, and it reflects real-world herd conditions. In this context, PS methods can be extremely useful as an inference tool for investigating causal effects. In addition, PS analysis can be implemented as a preliminary evaluation or a hypothesis generator for future randomized trials (if the trait analyzed allows randomization).

Genome-wide association and pathway-based analysis using latent variables related to milk protein composition and cheesemaking traits in dairy cattle.

The aim of this study was to perform genome-wide associations (GWAS) and gene-set enrichment analyses with protein composition and cheesemaking-related latent variables (factors; F) in a cohort of 1,011 Italian Brown Swiss cows. Factor analysis was applied to identify latent structures of 26 phenotypes related to bovine milk quantity and quality, protein fractions [αS1-, αS2-, ß-, and κ-casein (CN), ß-lactoglobulin, and α-lactalbumin (α-LA)], coagulation and curd firming at time t (CFt) measures, and cheese properties [cheese yield (%CY) and nutrients recovery in the curd] of individual cows. Ten orthogonal F were extracted, explaining 74% of the original variability. Factor 1%CY underlined the %CY characteristics, F2CFt was related to the CFt process parameters, F3Yield was considered as descriptor of milk and solids yield, whereas F4Cheese N underscored the presence of nitrogenous compounds (N) into the cheese. Four more F were related to the milk caseins (F5αS1-ß-CN, F7ß-κ-CN, F8αS2-CN, and F9αS1-CN-Ph) and 1 F was linked to the whey protein (F10α-LA); 1 F underlined the udder health status (F6Udder health). All cows were genotyped with the Illumina BovineSNP50 Bead Chip v.2 (Illumina Inc., San Diego, CA). Single marker regression GWAS were fitted. Gene-set enrichment analysis was run on GWAS results, using the Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway databases, to reveal ontologies or pathways associated with the F. All F but F3Yield showed significance in GWAS. Signals in 10 Bos taurus autosomes (BTA) were detected. High peaks on BTA6 (∼87 Mbp) were found for F6ß-κ-CN, F5αS1-ß-CN, and at the tail of BTA11 (∼104 Mbp) for F4Cheese N. Gene-set enrichment analyses showed significant results (false discovery rate at 5%) for F8αS2-CN, F1%CY, F4Cheese N, and F10α-LA. For F8αS2-CN, 33 Gene Ontology terms and 3 Kyoto Encyclopedia of Genes and Genomes categories were enriched, including terms related to ion transport and homeostasis, neuron function or part, and GnRH signaling pathway. Our results support the feasibility of factor analysis as a dimension reduction technique in genomic studies and evidenced a potential key role of αS2-CN in milk quality and composition.