Unbalanced metabolic status in the weeks after calving predisposes dairy cows to metabolic and infectious diseases. Blood glucose, IGF-I, non-esterified fatty acids (NEFA) and β-hydroxybutyrate (BHB) are used as indicators of the metabolic status of cows. This work aims to (1) evaluate the potential of milk mid-IR spectra to predict these blood components individually and (2) to evaluate the possibility of predicting the metabolic status of cows based on the clustering of these blood components. Blood samples were collected from 241 Holstein cows on six experimental farms, at days 14 and 35 after calving. Blood samples were analyzed by reference analysis and metabolic status was defined by k-means clustering (k=3) based on the four blood components. Milk mid-IR analyses were undertaken on different instruments and the spectra were harmonized into a common standardized format. Quantitative models predicting blood components were developed using partial least squares regression and discriminant models aiming to differentiate the metabolic status were developed with partial least squares discriminant analysis. Cross-validations were performed for both quantitative and discriminant models using four subsets randomly constituted. Blood glucose, IGF-I, NEFA and BHB were predicted with respective R 2 of calibration of 0.55, 0.69, 0.49 and 0.77, and R 2 of cross-validation of 0.44, 0.61, 0.39 and 0.70. Although these models were not able to provide precise quantitative values, they allow for screening of individual milk samples for high or low values. The clustering methodology led to the sharing out of the data set into three groups of cows representing healthy, moderately impacted and imbalanced metabolic status. The discriminant models allow to fairly classify the three groups, with a global percentage of correct classification up to 74%. When discriminating the cows with imbalanced metabolic status from cows with healthy and moderately impacted metabolic status, the models were able to distinguish imbalanced group with a global percentage of correct classification up to 92%. The performances were satisfactory considering the variables are not present in milk, and consequently predicted indirectly. This work showed the potential of milk mid-IR analysis to provide new metabolic status indicators based on individual blood components or a combination of these variables into a global status. Models have been developed within a standardized spectral format, and although robustness should preferably be improved with additional data integrating different geographic regions, diets and breeds, they constitute rapid, cost-effective and large-scale tools for management and breeding of dairy cows.

Improving nitrogen use efficiency (NUE) at both the individual cow and the herd level has become a key target in dairy production systems, for both environmental and economic reasons. Cost-effective and large-scale phenotyping methods are required to improve NUE through genetic selection and by feeding and management strategies. The aim of this study was to evaluate the possibility of using mid-infrared (MIR) spectra of milk to predict individual dairy cow NUE during early lactation. Data were collected from 129 Holstein cows, from calving until 50 d in milk, in 3 research herds (Denmark, Ireland, and the UK). In 2 of the herds, diets were designed to challenge cows metabolically, whereas a diet reflecting local management practices was offered in the third herd. Nitrogen intake (kg/d) and nitrogen excreted in milk (kg/d) were calculated daily. Nitrogen use efficiency was calculated as the ratio between nitrogen in milk and nitrogen intake, and expressed as a percentage. Individual daily values for NUE ranged from 9.7 to 81.7%, with an average of 36.9% and standard deviation of 10.4%. Milk MIR spectra were recorded twice weekly and were standardized into a common format to avoid bias between apparatus or sampling periods. Regression models predicting NUE using milk MIR spectra were developed on 1,034 observations using partial least squares or support vector machines regression methods. The models were then evaluated through (1) a cross-validation using 10 subsets, (2) a cow validation excluding 25% of the cows to be used as a validation set, and (3) a diet validation excluding each of the diets one by one to be used as validation sets. The best statistical performances were obtained when using the support vector machines method. Inclusion of milk yield and lactation number as predictors, in combination with the spectra, also improved the calibration. In cross-validation, the best model predicted NUE with a coefficient of determination of cross-validation of 0.74 and a relative error of 14%, which is suitable to discriminate between low- and high-NUE cows. When performing the cow validation, the relative error remained at 14%, and during the diet validation the relative error ranged from 12 to 34%. In the diet validation, the models showed a lack of robustness, demonstrating difficulties in predicting NUE for diets and for samples that were not represented in the calibration data set. Hence, a need exists to integrate more data in the models to cover a maximum of variability regarding breeds, diets, lactation stages, management practices, seasons, MIR instruments, and geographic regions. Although the model needs to be validated and improved for use in routine conditions, these preliminary results showed that it was possible to obtain information on NUE through milk MIR spectra. This could potentially allow large-scale predictions to aid both further genetic and genomic studies, and the development of farm management tools.

Both blood- and milk-based biomarkers have been analysed for decades in research settings, although often only in one herd, and without focus on the variation in the biomarkers that are specifically related to herd or diet. Biomarkers can be used to detect physiological imbalance and disease risk and may have a role in precision livestock farming (PLF). For use in PLF, it is important to quantify normal variation in specific biomarkers and the source of this variation. The objective of this study was to estimate the between- and within-herd variation in a number of blood metabolites (β-hydroxybutyrate (BHB), non-esterified fatty acids, glucose and serum IGF-1), milk metabolites (free glucose, glucose-6-phosphate, urea, isocitrate, BHB and uric acid), milk enzymes (lactate dehydrogenase and N-acetyl-β-D-glucosaminidase (NAGase)) and composite indicators for metabolic imbalances (Physiological Imbalance-index and energy balance), to help facilitate their adoption within PLF. Blood and milk were sampled from 234 Holstein dairy cows from 6 experimental herds, each in a different European country, and offered a total of 10 different diets. Blood was sampled on 2 occasions at approximately 14 days-in-milk (DIM) and 35 DIM. Milk samples were collected twice weekly (in total 2750 samples) from DIM 1 to 50. Multilevel random regression models were used to estimate the variance components and to calculate the intraclass correlations (ICCs). The ICCs for the milk metabolites, when adjusted for parity and DIM at sampling, demonstrated that between 12% (glucose-6-phosphate) and 46% (urea) of the variation in the metabolites’ levels could be associated with the herd-diet combination. Intraclass Correlations related to the herd-diet combination were generally higher for blood metabolites, from 17% (cholesterol) to approximately 46% (BHB and urea). The high ICCs for urea suggest that this biomarker can be used for monitoring on herd level. The low variance within cow for NAGase indicates that few samples would be needed to describe the status and potentially a general reference value could be used. The low ICC for most of the biomarkers and larger within cow variation emphasises that multiple samples would be needed - most likely on the individual cows - for making the biomarkers useful for monitoring. The majority of biomarkers were influenced by parity and DIM which indicate that these should be accounted for if the biomarker should be used for monitoring.

Direct measurements of methane (CH₄) from individual animals are difficult and expensive. Predictions based on proxies for CH₄ are a viable alternative. Most prediction models are based on multiple linear regressions (MLR) and predictor variables that are not routinely available in commercial farms, such as dry matter intake (DMI) and diet composition. The use of machine learning (ML) algorithms to predict CH₄ emissions from across-country heterogeneous data sets has not been reported. The objectives were to compare performances of ML ensemble algorithm random forest (RF) and MLR models in predicting CH₄ emissions from proxies in dairy cows, and assess effects of imputing missing data points on prediction accuracy. Data on CH₄ emissions and proxies for CH₄ from 20 herds were provided by 10 countries. The integrated data set contained 43,519 records from 3,483 cows, with 18.7% missing data points imputed using k-nearest neighbor imputation. Three data sets were created, 3k (no missing records), 21k (missing DMI imputed from milk, fat, protein, body weight), and 41k (missing DMI, milk fat, and protein records imputed). These data sets were used to test scenarios (with or without DMI, imputed vs. nonimputed DMI, milk fat, and protein), and prediction models (RF vs. MLR). Model predictive ability was evaluated within and between herds through 10-fold cross-validation. Prediction accuracy was measured as correlation between observed and predicted CH₄, root mean squared error (RMSE) and mean normalized discounted cumulative gain (NDCG). Inclusion of DMI in the model improved within and between-herd prediction accuracy to 0.77 (RMSE = 23.3%) and 0.58 (RMSE = 31.9%) in RF and to 0.50 (RMSE = 0.327) and 0.13 (RMSE = 42.71) in MLR, respectively than when DMI was not included in the predictive model. When missing DMI records were imputed, within and between-herd accuracy increased to 0.84 (RMSE = 18.5%) and 0.63 (RMSE = 29.9%), respectively. In all scenarios, RF models out-performed MLR models. Results suggest routinely measured variables from dairy farms can be used in developing globally robust prediction models for CH₄ if coupled with state-of-the-art techniques for imputation and advanced ML algorithms for predictive modeling.