Accurately identifying pregnancy status is imperative for a profitable dairy enterprise. Mid-infrared (MIR) spectroscopy is routinely used to determine fat and protein concentrations in milk samples. Mid-infrared spectra have successfully been used to predict other economically important traits, including fatty acid content, mineral content, body energy status, lactoferrin, feed intake, and methane emissions. Machine learning has been used in a variety of fields to find patterns in vast quantities of data. This study aims to use deep learning, a sub-branch of machine learning, to establish pregnancy status from routinely collected milk MIR spectral data. Milk spectral data were obtained from National Milk Records (Chippenham, UK), who collect large volumes of data continuously on a monthly basis. Two approaches were followed: using genetic algorithms for feature selection and network design (model 1), and transfer learning with a pretrained DenseNet model (model 2). Feature selection in model 1 showed that the number of wave points in MIR data could be reduced from 1,060 to 196 wave points. The trained model converged after 162 epochs with validation accuracy and loss of 0.89 and 0.18, respectively. Although the accuracy was sufficiently high, the loss (in terms of predicting only 2 labels) was considered too high and suggested that the model would not be robust enough to apply to industry. Model 2 was trained in 2 stages of 100 epochs each with spectral data converted to gray-scale images and resulted in accuracy and loss of 0.97 and 0.08, respectively. Inspection on inference data showed prediction sensitivity of 0.89, specificity of 0.86, and prediction accuracy of 0.88. Results indicate that milk MIR data contains features relating to pregnancy status and the underlying metabolic changes in dairy cows, and such features can be identified by means of deep learning. Prediction equations from trained mod-els can be used to alert farmers of nonviable pregnancies as well as to verify conception dates.

BackgroundGyrodactylus salaris Malmberg, 1957 has had a devastating impact on wild Norwegian stocks of Atlantic salmon Salmo salar L., and it is the only Office International des Epizooties (OIE) listed parasitic pathogen of fish. The UK is presently recognised as G. salaris-free, and management plans for its containment and control are currently based on Scandinavian studies. The current study investigates the susceptibility of British salmonids to G. salaris, and determines whether, given the host isolation since the last glaciation and potential genetic differences, the populations under test would exhibit different levels of susceptibility, as illustrated by the parasite infection trajectory over time, from their Scandinavian counterparts.MethodsPopulations of S. salar, brown trout Salmo trutta L., and grayling Thymallus thymallus (L.), raised from wild stock in UK government hatcheries, were flown to Norway and experimentally challenged with a known pathogenic strain of G. salaris. Each fish was lightly anaesthetised and marked with a unique tattoo for individual parasite counting. A single Norwegian population of S. salar from the River Lærdalselva was used as a control. Parasite numbers were assessed every seven days until day 48 and then every 14 days.ResultsGyrodactylus salaris regularly leads to high mortalities on infected juveniles S. salar. The number of G. salaris on British S. salar rose exponentially until the experiment was terminated at 33 days due to fish welfare concerns. The numbers of parasites on S. trutta and T. thymallus increased sharply, reaching a peak of infection on days 12 and 19 post-infection respectively, before declining to a constant low level of infection until the termination of the experiment at 110 days.ConclusionsThe ability of S. trutta and T. thymallus to carry an infection for long periods increases the window of exposure for these two hosts and the potential transfer of G. salaris to other susceptible hosts. This study demonstrates that G. salaris can persist on S. trutta for longer periods than previously thought, and that the role that S. trutta could play in disseminating G. salaris needs to be considered carefully and factored into management plans and epidemics across Europe.

Data collected from an experimental Holstein-Friesian research herd were used to determine genetic and phenotypic parameters of innate and adaptive cellular immune-associated traits. Relationships between immune-associated traits and production, health, and fertility traits were also investigated. Repeated blood leukocyte records were analyzed in 546 cows for 9 cellular immune-associated traits, including percent T cell subsets, B cells, NK cells, and granulocytes. Variance components were estimated by univariate analysis. Heritability estimates were obtained for all 9 traits, the highest of which were observed in the T cell subsets percent CD4, percent CD8, CD4:CD8 ratio, and percent NKp46 cells (0.46, 0.41, 0.43 and 0.42, respectively), with between-individual variation accounting for 59 to 81% of total phenotypic variance. Associations between immune-associated traits and production, health, and fertility traits were investigated with bivariate analyses. Strong genetic correlations were observed between percent NKp46 and stillbirth rate (0.61), and lameness episodes and percent CD8 (-0.51). Regarding production traits, the strongest relationships were between CD4:CD8 ratio and weight phenotypes (-0.52 for live weight; -0.51 for empty body weight). Associations between feed conversion traits and immune-associated traits were also observed. Our results provide evidence that cellular immune-associated traits are heritable and repeatable, and the noticeable variation between animals would permit selection for altered trait values, particularly in the case of the T cell subsets. The associations we observed between immune-associated, health, fertility, and production traits suggest that genetic selection for cellular immune-associated traits could provide a useful tool in improving animal health, fitness, and fertility.

Bovine tuberculosis (bTB) is a zoonotic disease in cattle that is transmissible to humans, distributed worldwide, and considered endemic throughout much of England and Wales. Mid-infrared (MIR) analysis of milk is used routinely to predict fat and protein concentration, and is also a robust predictor of several other economically important traits including individual fatty acids and body energy. This study predicted bTB status of UK dairy cows using their MIR spectral profiles collected as part of routine milk recording. Bovine tuberculosis data were collected as part of the national bTB testing program for Scotland, England, and Wales; these data provided information from over 40,500 bTB herd breakdowns. Corresponding individual cow life-history data were also available and provided information on births, movements, and deaths of all cows in the study. Data relating to single intradermal comparative cervical tuberculin (SICCT) skin-test results, culture, slaughter status, and presence of lesions were combined to create a binary bTB phenotype labeled 0 to represent nonresponders (i.e., healthy cows) and 1 to represent responders (i.e., bTB-affected cows). Contemporaneous individual milk MIR spectral data were collected as part of monthly routine milk recording and matched to bTB status of individual animals on the single intradermal comparative cervical tuberculin test date (±15 d). Deep learning, a sub-branch of machine learning, was used to train artificial neural networks and develop a prediction pipeline for subsequent use in national herds as part of routine milk recording. Spectra were first converted to 53 × 20-pixel PNG images, then used to train a deep convolutional neural network. Deep convolutional neural networks resulted in a bTB prediction accuracy (i.e., the number of correct predictions divided by the total number of predictions) of 71% after training for 278 epochs. This was accompanied by both a low validation loss (0.71) and moderate sensitivity and specificity (0.79 and 0.65, respectively). To balance data in each class, additional training data were synthesized using the synthetic minority over sampling technique. Accuracy was further increased to 95% (after 295 epochs), with corresponding validation loss minimized (0.26), when synthesized data were included during training of the network. Sensitivity and specificity also saw a 1.22-and 1.45-fold increase to 0.96 and 0.94, respectively, when synthesized data were included during training. We believe this study to be the first of its kind to predict bTB status from milk MIR spectral data. We also believe it to be the first study to use milk MIR spectral data to predict a disease phenotype, and posit that the automated prediction of bTB status at routine milk recording could provide farmers with a robust tool that enables them to make early management decisions on potential reactor cows, and thus help slow the spread of bTB.

noted between milk and serum Ca (0.17), Mo (0.19), and Na (−0.79). Additional multivariate analyses between measures within sample type (i.e., milk or serum) revealed significant positive associations, both phenotypic and genetic, between some of the elements. In milk, Se was genetically correlated with Ca (0.63), Mg (0.59), Mn (0.40), P (0.53), and Zn (0.52), whereas in serum, V showed strong genetic associations with Cd (0.71), Ca (0.53), Mn (0.63), Mo (0.57), P (0.42), K (0.45), and Hg (−0.44). These results provide evidence that element concentrations in milk and blood of dairy cows are significantly influenced by both diet and genetics and demonstrate the potential for genetic selection and dietary manipulation to alter nutrient concentration to improve both cow health and the healthfulness of milk for human consumption.