Energy profiling of dairy cows from routine milk mid-infrared analysis

The balance of body energy within and across lactations can have health and fertility consequences for the dairy cow. This study aimed to create a large calibration data set of dairy cow body energy traits across the cow's productive life, with concurrent milk mid-infrared (MIR) spectral data, to generate a prediction tool for use in commercial dairy herds. Detailed phenotypic data from 1,101 Holstein Friesian cows from the Langhill research herd (SRUC, Scotland) were used to generate energy balance (EB) and effective energy intake (EI), both in megajoules per day. Pretreatment of spectral data involved standardization to account for drift over time and machine. Body energy estimates were aligned with their spectral data to generate a prediction of these traits based on milk MIR spectroscopy. After data edits, partial least squares analysis generated prediction equations with a coefficient of determination from split sample 10-fold cross validation of 0.77 and 0.75 for EB and EI, respectively. These prediction equations were applied to national milk MIR spectra on over 11 million animal test dates (January 2013 to December 2016) from 4,453 farms. The predictions generated from these were subject to phenotypic analyses with a fixed regression model highlighting differences between the main dairy breeds in terms of energy traits. Genetic analyses generated heritability estimates for EB and EI ranging from 0.12 to 0.17 and 0.13 to 0.15, respectively. This study shows that MIR-based predictions from routinely collected national data can be used to generate predictions of dairy cow energy turnover profiles for both animal management and genetic improvement of such difficult and expensive-to-record traits.     

Mid-infrared spectrometry of milk as a predictor of energy intake and efficiency in lactating dairy cows

Interest is increasing in the feed intake complex of individual dairy cows, both for management and animal breeding. However, energy intake data on an individual-cow basis are not routinely available. The objective of the present study was to quantify the ability of routinely undertaken mid-infrared (MIR) spectroscopy analysis of individual cow milk samples to predict individual cow energy intake and efficiency. Feed efficiency in the present study was described by residual feed intake (RFI), which is the difference between actual energy intake and energy used (e.g., milk production, maintenance, and body tissue anabolism) or supplied from body tissue mobilization. A total of 1,535 records for energy intake, RFI, and milk MIR spectral data were available from an Irish research herd across 36 different test days from 535 lactations on 378 cows. Partial least squares regression analyses were used to relate the milk MIR spectral data to either energy intake or efficiency. The coefficient of correlation (R_EX) of models to predict RFI across lactation ranged from 0.48 to 0.60 in an external validation data set; the predictive ability was, however, strongest (R_EX = 0.65) in early lactation (<60 d in milk). The inclusion of milk yield as a predictor variable improved the accuracy of predicting energy intake across lactation (R_EX = 0.70). The correlation between measured RFI and measured energy balance across lactation was 0.85, whereas the correlation between RFI and energy balance, both predicted from the MIR spectrum, was 0.65. Milk MIR spectral data are routinely generated for individual cows throughout lactation and, therefore, the prediction equations developed in the present study can be immediately (and retrospectively where MIR spectral data have been stored) applied to predict energy intake and efficiency to aid in management and breeding decisions.     

Validation of mid-infrared spectrometry in milk for predicting body energy status in Holstein-Friesian cows

Cow energy balance is known to be associated with cow health and fertility; therefore, routine access to data on energy balance can be useful in both management and breeding decisions to improve cow performance. The objective of this study was to determine if individual cow milk mid-infrared spectra (MIR) could be useful to predict cow energy balance across contrasting production systems. Direct energy balance was calculated as the differential between energy intake and energy output in milk and maintenance (maintenance was predicted using body weight). Body energy content was calculated from (change in) body weight and body condition score. Following editing, 2,992 morning, 2,742 midday, and 2,989 evening milk MIR records from 564 lactations on 337 Scottish cows, managed in a confinement system on 1 of 2 diets, were available. An additional 844 morning and 820 evening milk spectral records from 338 lactations on 244 Irish cows offered a predominantly grazed grass diet were also available. Equations were developed to predict body energy status using the milk spectral data and milk yield as predictor variables. Several different approaches were used to test the robustness of the equations calibrated in one data set and validated in another. The analyses clearly showed that the variation in the validation data set must be represented in the calibration data set. The accuracy (i.e., square root of the coefficient of multiple determinations) of predicting, from MIR, direct energy balance, body energy content, and energy intake was 0.47 to 0.69, 0.51 to 0.56, and 0.76 to 0.80, respectively. This highlights the ability of milk MIR to predict body energy balance, energy content, and energy intake with reasonable accuracy. Very high accuracy, however, was not expected, given the likely random errors in the calculation of these energy status traits using field data.     

Predicting pregnancy status from mid-infrared spectroscopy in dairy cow milk using deep learning

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 models can be used to alert farmers of nonviable pregnancies as well as to verify conception dates.     

Prediction of blood metabolites from milk mid-infrared spectra in early-lactation cows

Dairy cows commonly experience an unbalanced energy status in early lactation, and this condition can lead to the onset of several metabolic disorders. Blood metabolic profile testing is a valid tool to monitor and detect the most common early lactation disorders, but blood sampling and analysis are time-consuming and expensive, and the procedure is invasive and stressful for the cows. Mid-infrared (MIR) spectroscopy is routinely used to analyze milk composition, being a cost-effective and nondestructive method. The present study aimed to assess the feasibility of using routine milk MIR spectra for the prediction of main blood metabolites in dairy cows, and to investigate associations between measured blood metabolites and milk traits. Twenty herds of Holstein Friesian, Brown Swiss, or Simmental cows located in Northeast Italy were visited 1 to 4 times between December 2017 and June 2018, and blood and milk samples were collected from all lactating cows within 35 d in milk. Concentrations of main blood metabolites and milk MIR spectra were recorded from 295 blood and milk samples and used to develop prediction models for blood metabolic traits through backward interval partial least squares analysis. Blood β-hydroxybutyrate (BHB), urea, and nonesterified fatty acids were the most predictable traits, with coefficients of determination of 0.63, 0.58, and 0.52, respectively. On the contrary, predictive performance for blood glucose, triglycerides, cholesterol, glutamic oxaloacetic transaminase, and glutamic pyruvic transaminase were not accurate. Associations of blood BHB and urea with their respective contents in milk were moderate to strong, whereas all other correlations were weak. Predicted blood BHB showed an improved performance in detecting cows with hyperketonemia (blood BHB ≥ 1.2 mmol/L), compared with commercial calibration equation for milk BHB. Results highlighted the opportunity of using milk MIR spectra to predict blood metabolites and thus to collect routine information on the metabolic status of early-lactation cows at a population level.     

The relationship between body energy traits and production and fitness traits in first-lactation dairy cows

Daily breeding values for various body energy traits were estimated from nationally recorded linear type information using random regression techniques. These traits included predicted live weight, growth rate (GR), and several measures of body energy during first lactation. The relationship between the individual linear type traits, body energy traits, and other production and fitness traits was examined by estimating the approximate genetic correlations (and standard errors) from adjusted predicted transmitting ability correlations. The relationship between fitness and body traits was further examined by calculating the partial correlation between the pairs of traits at a constant milk yield. Daily sire solutions for type traits showed that there were differences in how cows changed body shape across lactation. On average, cows lost body energy in early lactation but regained it by the end of the lactation. However, there was large variation, with the daughters of some sires still in negative energy balance by the end of lactation. The estimates of genetic correlation of linear type traits with production and fitness traits agreed with previous linear point estimates. The fertility traits were correlated across lactation with all linear type traits studied and indicated that bigger animals tend to have poorer fertility. There was an unfavorable correlation between production and GR; therefore, higher producing cows were more likely to have lower GR early in first lactation. The results also showed that higher producing animals were likely to loose body energy during the peak of first lactation. This is mirrored in the relationship of early lactation energy balance with somatic cell count and longevity, indicating that peak lactation is a critical time in terms of metabolic pressures on the dairy cow.     

Evaluation of five lactation curve models fitted for fat:protein ratio of milk and daily energy balance

Selection for milk yield increases the metabolic load of dairy cows. The fat:protein ratio of milk (FPR) could serve as a measure of the energy balance status and might be used as a selection criterion to improve metabolic stability. The fit of different fixed and random regression models describing FPR and daily energy balance was tested to establish appropriate models for further genetic analyses. In addition, the relationship between both traits was evaluated for the best fitting model. Data were collected on a dairy research farm running a bull dam performance test. Energy balance was calculated using information on milk yield, feed intake per day, and live weight. Weekly FPR measurements were available. Three data sets were created containing records of 577 primiparous cows with observations from lactation d 11 to 180 as well as records of 613 primiparous cows and 96 multiparous cows with observations from lactation d 11 to 305. Five well-established parametric functions of days in milk (Ali and Schaeffer, Guo and Swalve, Wilmink, Legendre polynomials of third and fourth degree) were chosen for modeling the lactation curves. Evaluation of goodness of fit was based on the corrected Akaike information criterion, the Bayesian information criterion, correlation between the real observation and the estimated value, and on inspection of the residuals plotted against days in milk. The best model was chosen for estimation of correlations between both traits at different lactation stages. Random regression models were superior compared with the fixed regression models. In general, the Ali and Schaeffer function appeared most suitable for modeling both the fixed and the random regression part of the mixed model. The FPR is greatest in the initial lactation period when energy deficit is most pronounced. Energy balance stabilizes at the same point as the decrease in FPR stops. The inverted patterns indicate a causal relationship between the 2 traits. A common pattern was also observed for repeatabilities of both traits, with repeatabilities being largest at the beginning of lactation. Additionally, correlations between cow effects were closest at the beginning of lactation (r_c = −0.43). The results support the hypothesis that FPR can serve as a suitable indicator for energy status, at least during the most metabolically stressful stage of lactation.     

Correlations of feed intake predicted with milk infrared spectra and breeding values in the Dutch Holstein population

Feed is a major cost in dairy production, and substantial genetic variation in feed efficiency exists between cows. Therefore, breeders aim to improve feed efficiency of dairy cattle. However, phenotypic data on individual feed intake on commercial farms is scarce, and accurate measurements are very costly. Several studies have shown that information from Fourier-transformed infrared spectra of milk samples (milk infrared, milk IR) can be used to predict phenotypes such as energy balance and energy intake, but this is usually based on small data sets obtained under experimental circumstances. The added value of information from milk IR spectra for estimation of breeding values is unknown. The objectives of this study were (1) to develop prediction equations for dry matter intake (DMI) and residual DMI (rDMI) from milk IR spectra; (2) to apply these for a data set of milk IR spectra from commercial Dutch dairy farms; (3) to estimate genetic parameters for these traits; and (4) to estimate correlations between these predictions and other traits in the breeding goal. We used data from feeding trials where individual feed intake was recorded daily and for which milk IR spectra were determined weekly to develop prediction equations for DMI and rDMI with partial least squares regression. This data set contained over 7,600 weekly averaged DMI records linked with milk IR spectra from 271 cows. The equations were applied for a data set with test day information from 676 Dutch dairy herds with 621,567 records of 78,488 cows. Both milk IR-predicted DMI and rDMI were analyzed with an animal model to obtain genetic parameters and sire effect estimates that could be correlated with breeding values. A partial least squares regression model with 10 components from the milk IR spectra explained around 25% of DMI variation and less than 10% of rDMI variation in the validation set. Nearly all variation in the milk IR spectra was captured by 7 components; additional components contributed marginally to the spectral variation but decreased prediction errors for both traits. Accuracies of predictions of DMI and rDMI from milk IR spectra for a large feeding experiment were 0.47 and 0.26 on average, respectively, with small differences between ration treatments (ranging from 0.43 to 0.55 and from 0.21 to 0.34, respectively) and among lactation stages (ranging from 0.24 to 0.59 and from 0.13 to 0.36, respectively), with the highest prediction accuracies in early lactation. The estimated heritabilities for predicted DMI and rDMI were 0.3 and 0.4, respectively, which suggests genetic potential for both predicted traits. The correlations of sire estimates for milk IR-predicted DMI with official Dutch breeding values were strongest with milk production (0.33), longevity (0.26), and fertility (?0.27), indicating that cows that eat more produce more, live longer, and have poorer fertility. The correlations of sire estimates for predicted DMI and rDMI with the official breeding values for DMI were low (0.14 and 0.03, respectively). This implies that the added value of including milk IR-predicted DMI information in the estimation procedure of breeding values for DMI would be considered insufficient for practical application.     

Body and milk traits as indicators of dairy cow energy status in early lactation

The inclusion of feed intake and efficiency traits in dairy cow breeding goals can lead to increased risk of metabolic stress. An easy and inexpensive way to monitor postpartum energy status (ES) of cows is therefore needed. Cows' ES can be estimated by calculating the energy balance from energy intake and output and predicted by indicator traits such as change in body weight (ΔBW), change in body condition score (ΔBCS), milk fat:protein ratio (FPR), or milk fatty acid (FA) composition. In this study, we used blood plasma nonesterified fatty acids (NEFA) concentration as a biomarker for ES. We determined associations between NEFA concentration and ES indicators and evaluated the usefulness of body and milk traits alone, or together, in predicting ES of the cow. Data were collected from 2 research herds during 2013 to 2016 and included 137 Nordic Red dairy cows, all of which had a first lactation and 59 of which also had a second lactation. The data included daily body weight, milk yield, and feed intake and monthly BCS. Plasma samples for NEFA were collected twice in lactation wk 2 and 3 and once in wk 20. Milk samples for analysis of fat, protein, lactose, and FA concentrations were taken on the blood sampling days. Plasma NEFA concentration was higher in lactation wk 2 and 3 than in wk 20 (0.56 ± 0.30, 0.43 ± 0.22, and 0.13 ± 0.06 mmol/L, respectively; all means ± standard deviation). Among individual indicators, C18:1 cis-9 and the sum of C18:1 in milk had the highest correlations (r = 0.73) with NEFA. Seven multiple linear regression models for NEFA prediction were developed using stepwise selection. Of the models that included milk traits (other than milk FA) as well as body traits, the best fit was achieved by a model with milk yield, FPR, ΔBW, ΔBCS, FPR × ΔBW, and days in milk. The model resulted in a cross-validation coefficient of determination (R²cv) of 0.51 and a root mean squared error (RMSE) of 0.196 mmol/L. When only milk FA concentrations were considered in the model, NEFA prediction was more accurate using measurements from evening milk than from morning milk (R²cv = 0.61 vs. 0.53). The best model with milk traits contained FPR, C10:0, C14:0, C18:1 cis-9, C18:1 cis-9 × C14:0, and days in milk (R²cv = 0.62; RMSE = 0.177 mmol/L). The most advanced model using both milk and body traits gave a slightly better fit than the model with only milk traits (R²cv = 0.63; RMSE = 0.176 mmol/L). Our findings indicate that ES of cows in early lactation can be monitored with moderately high accuracy by routine milk measurements.     

Milk mid-infrared spectral data as a tool to predict feed intake in lactating Norwegian Red dairy cows

Mid-infrared (MIR) spectroscopy of milk was used to predict dry matter intake (DMI) and net energy intake (NEI) in 160 lactating Norwegian Red dairy cows. A total of 857 observations were used in leave-one-out cross-validation and external validation to develop and validate prediction equations using 5 different models. Predictions were performed using (multiple) linear regression, partial least squares (PLS) regression, or best linear unbiased prediction (BLUP) methods. Linear regression was implemented using just milk yield (MY) or fat, protein, and lactose concentration in milk (Mcont) or using MY together with body weight (BW) as predictors of intake. The PLS and BLUP methods were implemented using just the MIR spectral information or using the MIR together with Mcont, MY, BW, or NEI from concentrate (NEIconc). When using BLUP, the MIR spectral wavelengths were always treated as random effects, whereas Mcont, MY, BW, and NEIconc were considered to be fixed effects. Accuracy of prediction (R) was defined as the correlation between the predicted and observed feed intake test-day records. When using the linear regression method, the greatest R of predicting DMI (0.54) and NEI (0.60) in the external validation was achieved when the model included both MY and BW. When using PLS, the greatest R of predicting DMI (0.54) and NEI (0.65) in the external validation data set was achieved when using both BW and MY as predictors in combination with the MIR spectra. When using BLUP, the greatest R of predicting DMI (0.54) in the external validation was when using MY together with the MIR spectra. The greatest R of predicting NEI (0.65) in the external validation using BLUP was achieved when the model included both BW and MY in combination with the MIR spectra or when the model included both NEIconc and MY in combination with MIR spectra. However, although the linear regression coefficients of actual on predicted values for DMI and NEI were not different from unity when using PLS, they were less than unity for some of the models developed using BLUP. This study shows that MIR spectral data can be used to predict NEI as a measure of feed intake in Norwegian Red dairy cattle and that the accuracy is augmented if additional, often available data are also included in the prediction model.     

Mid-infrared spectroscopy predictions as indicator traits in breeding programs for enhanced coagulation properties of milk

The aims of this study were to investigate variation of milk coagulation property (MCP) measures and their predictions obtained by mid-infrared spectroscopy (MIR), to investigate the genetic relationship between measures of MCP and MIR predictions, and to estimate the expected response from a breeding program focusing on the enhancement of MCP using MIR predictions as indicator traits. Individual milk samples were collected from 1,200 Brown Swiss cows (progeny of 50 artificial insemination sires) reared in 30 herds located in northern Italy. Rennet coagulation time (RCT, min) and curd firmness (a₃₀, mm) were measured using a computerized renneting meter. The MIR data were recorded over the spectral range of 4,000 to 900 cm⁻¹. Prediction models for RCT and a₃₀ based on MIR spectra were developed using partial least squares regression. A cross-validation procedure was carried out. The procedure involved the partition of available data into 2 subsets: a calibration subset and a test subset. The calibration subset was used to develop a calibration equation able to predict individual MCP phenotypes using MIR spectra. The test subset was used to validate the calibration equation and to estimate heritabilities and genetic correlations for measured MCP and their predictions obtained from MIR spectra and the calibration equation. Point estimates of heritability ranged from 0.30 to 0.34 and from 0.22 to 0.24 for RCT and a₃₀, respectively. Heritability estimates for MCP predictions were larger than those obtained for measured MCP. Estimated genetic correlations between measures and predictions of RCT were very high and ranged from 0.91 to 0.96. Estimates of the genetic correlation between measures and predictions of a₃₀ were large and ranged from 0.71 to 0.87. Predictions of MCP provided by MIR techniques can be proposed as indicator traits for the genetic enhancement of MCP. The expected response of RCT and a₃₀ ensured by the selection using MIR predictions as indicator traits was equal to or slightly less than the response achievable through a single measurement of these traits. Breeding strategies for the enhancement of MCP based on MIR predictions as indicator traits could be easily and immediately implemented for dairy cattle populations where routine acquisition of spectra from individual milk samples is already performed.     

The relationships between early lactation energy status indicators and endocrine fertility traits in dairy cows

The relationships between dairy cow milk-based energy status (ES) indicators and fertility traits were studied during periods 8 to 21, 22 to 35, 36 to 49, and 50 to 63 d in milk. Commencement of luteal activity (C-LA) and interval from calving to the first heat (CFH), based on frequent measurements of progesterone by the management tool Herd Navigator (DeLaval), were used as fertility traits. Energy status indicator traits were milk β-hydroxybutyrate (BHB) concentration provided by Herd Navigator and milk fat:protein ratio, concentration of C18:1 cis-9, the ratio of fatty acids (FA) C18:1 cis-9 and C10:0 in test-day milk samples, and predicted plasma concentration of nonesterified fatty acids (NEFA) on test days. Plasma NEFA predictions were based either directly on milk mid-infrared spectra (MIR) or on milk fatty acids based on MIR spectra (NEFA_mir and NEFA_fa, respectively). The average (standard deviation) C-LA was 39.3 (±16.6) days, and the average CFH was 50.7 (±17.2) days. The correlations between fertility traits and ES indicators tended to be higher for multiparous (r < 0.28) than for primiparous (r < 0.16) cows. All correlations were lower in the last period than in the other periods. In period 1, correlations of C-LA with NEFA_fa and BHB, respectively, were 0.15 and 0.14 for primiparous and 0.26 and 0.22 for multiparous cows. The associations between fertility traits and ES indicators indicated that negative ES during the first weeks postpartum may delay the onset of luteal activity. Milk FPR was not as good an indicator for cow ES as other indicators. According to these findings, predictions of plasma NEFA and milk FA based on milk MIR spectra of routine test-day samples and the frequent measurement of milk BHB by Herd Navigator gave equally good predictions of cow ES during the first weeks of lactation. Our results indicate that routinely measured milk traits can be used for ES evaluation in early lactation.     

Metabolic profiling of early-lactation dairy cows using milk mid-infrared spectra

Metabolic disorders in early lactation have negative effects on dairy cow health and farm profitability. One method for monitoring the metabolic status of cows is metabolic profiling, which uses associations between the concentrations of several metabolites in serum and the presence of metabolic disorders. In this cross-sectional study, we investigated the use of mid-infrared (MIR) spectroscopy of milk for predicting the concentrations of these metabolites in serum. Between July and October 2017, serum samples were taken from 773 early-lactation Holstein Friesian cows located on 4 farms in the Gippsland region of southeastern Victoria, Australia, on the same day as milk recording. The concentrations in sera of β-hydroxybutyrate (BHB), fatty acids, urea, Ca, Mg, albumin, and globulins were measured by a commercial diagnostic laboratory. Optimal concentration ranges for each of the 7 metabolites were obtained from the literature. Animals were classified as being either affected or unaffected with metabolic disturbances based on these ranges. Milk samples were analyzed by MIR spectroscopy. The relationships between serum metabolite concentrations and MIR spectra were investigated using partial least squares regression. Partial least squares discriminant analyses (PLS-DA) were used to classify animals as being affected or not affected with metabolic disorders. Calibration equations were constructed using data from a randomly selected subset of cows (n = 579). Data from the remaining cows (n = 194) were used for validation. The coefficient of determination (R²) of serum BHB, fatty acids, and urea predictions were 0.48, 0.61, and 0.90, respectively. Predictions of Ca, Mg, albumin, and globulin concentrations were poor (0.06 ≤ R² ≤ 0.17). The PLS-DA models could predict elevated fatty acid and urea concentrations with an accuracy of approximately 77 and 94%, respectively. A second independent validation data set was assembled in March 2018, comprising blood and milk samples taken from 105 autumn-calving cows of various breeds. The accuracies of BHB and fatty acid predictions were similar to those obtained using the first validation data set. The PLS-DA results were difficult to interpret due to the low prevalence of metabolic disorders in the data set. Our results demonstrate that MIR spectroscopy of milk shows promise for predicting the concentration of BHB, fatty acids, and urea in serum; however, more data are needed to improve prediction accuracies.     

Phenotyping metabolic status of dairy cows using clustering of time profiles of energy balance peripartum

Due to a combination of a relatively low energy intake and a high demand of energy required for milk production, dairy cows experience a negative energy balance (EB) at the start of lactation. This energy deficit causes body weight reduction and an increased risk for metabolic diseases. Severity and length of negative EB can differ among cows. Peripartum time profiles of EB for dairy cows are not described yet in the literature. Creating EB-derived time profiles with corresponding metabolic status and disease treatments could improve understanding the relationship between EB and metabolic status, as well as enhance identification of cows at risk for compromised metabolic status. In this research we propose a novel method to cluster EB time series and examine associated metabolic status and disease treatments of dairy cows in the peripartum period. In this study, data of 3 earlier experiments were merged and examined. Four dairy cow clusters for time profiles of EB from wk ?3 until +7 relative to calving were generated by the global alignment kernel algorithm. For each cluster, mean of body weight prepartum was distinguishable, indicating this might be a possible on-farm biomarker for the peripartum EB profile. Moreover, cows with severe EB drop postpartum were more treated for milk fever and had high plasma nonesterified fatty acids and β-hydroxybutyrate concentration, and low IGF-1, insulin, and glucose concentration in the first 7 wk of lactation. Overall, this study demonstrated that cows can be clustered based on EB time profiles and that characteristics such as prepartum body weight, and postpartum nonesterified fatty acids and glucose concentration are promising biomarkers to identify the time profile of EB and potentially the risk for metabolic diseases.     

Prediction of dry matter intake throughout lactation in a dynamic model of dairy cow performance

In the dynamic modeling of dairy cow performance over a full lactation, the difference between net energy intake and net energy used for maintenance, growth, and output in milk accumulates in body reserves. A simple dynamic model of net energy balance was constructed to select, out of some common dry matter intake (DMI) prediction equations, the one that resulted in a minimum cumulative bias in body energy deposition. Dry matter intake was predicted using the Cornell Net Carbohydrate and Protein System, Agricultural Research Council, or National Research Council (NRC) DMI equations from body weight (BW) and predicted fat-corrected milk yield. The instantaneous BW of cows at progressive weeks of lactation was simulated as the numerical integral of the BW change obtained from the predicted net energy balance. Predicted DMI and BW from each DMI equation, using either of 2 equations to describe maintenance energy expenditures, were compared statistically against observed data from 21 herd average published full lactation data sets. All DMI equations underpredicted BW and DMI, but the NRC DMI equation resulted in the minimum cumulative error in predicted BW and DMI. As a general solution to prevent predicted BW from deviating substantially over time from the observed BW, a lipostatic feedback mechanism was integrated into the NRC DMI equation as a 2-parameter linear function of the relative size of simulated body reserves and week of lactation. Residual sum of squares was reduced on average by 52% for BW predictions and by 41% for DMI predictions by inclusion of the negative feedback with parameters taken from the average of all 21 least squares fits. Similarly, root mean square prediction error (%) was reduced by 30% on average for BW predictions and by 23% for DMI predictions. Inclusion of a feedback of energy reserves onto predicted DMI, simulating lipostatic regulation of BW, solved the problem of final BW deviation within a dynamic model and improved its DMI prediction to a satisfactory level.     

On the use of milk composition measures to predict the energy balance of dairy cows

Milk composition varies with energy status and was proposed for measuring energy balance on-farm, but the accuracy of prediction using monthly samples is not high. With automated sampling and inline milk analysis, a much higher measurement frequency is possible, and thus improved accuracy of energy balance determination may be expected. Energy balance was evaluated using data in which milk composition was measured at each milking. Three breeds (Danish Holstein, Danish Red, and Jerseys) of cows (623 lactations from 299 cows) in parities 1, 2, and 3+ were used. Data were smoothed using a rolling local regression. Energy balance (EBal) was calculated from changes in body reserves (body weight and body condition score). The relationship between EBal and milk measures was quantified by partial least squares regression (PLS) using group means data. For each day in lactation, the within-breed and parity mean EBal and mean milk measures were used. Further PLS was done using the individual cow data. The initial PLS models included 25 combinations of milk measures allowing a range of nonlinear effects. These combinations were as follows: days in milk (DIM); DIM raised to the powers 2, 3, and 4; milk yield; fat content; protein content; lactose content; fat yield; protein yield; lactose yield; fat:protein ratio; fat:lactose ratio; protein:lactose ratio; and milk yield:lactose ratio, together with 10 “diff()” variables. These variables are the current minus the previous value of the milk measure in question. Using group means data, a very high proportion (96%) of the variability in EBal was explained by the PLS model. A reduced model with only 6 variables explained 94% of the variation in EBal. This model had a prediction error of 3.82 MJ/d; the 25-variable model had a prediction error of 3.11 MJ/d. When using individual rather than group means data, the PLS prediction error was 17.3 MJ/d. In conclusion, the mean Ebal of different parities of Holstein, Danish Red, and Jersey cows can be predicted throughout lactation using 1 common equation based on DIM, milk yield, milk fat, and milk protein measures.     

Prediction of metabolic status of dairy cows in early lactation with on-farm cow data and machine learning algorithms

Metabolic status of dairy cows in early lactation can be evaluated using the concentrations of plasma β-hydroxybutyrate (BHB), free fatty acids (FFA), glucose, insulin, and insulin-like growth factor 1 (IGF-1). These plasma metabolites and metabolic hormones, however, are difficult to measure on farm. Instead, easily obtained on-farm cow data, such as milk production traits, have the potential to predict metabolic status. Here we aimed (1) to investigate whether metabolic status of individual cows in early lactation could be clustered based on their plasma values and (2) to evaluate machine learning algorithms to predict metabolic status using on-farm cow data. Through lactation wk 1 to 7, plasma metabolites and metabolic hormones of 334 cows were measured weekly and used to cluster each cow into 1 of 3 clusters per week. The cluster with the greatest plasma BHB and FFA and the lowest plasma glucose, insulin, and IGF-1 was defined as poor metabolic status; the cluster with the lowest plasma BHB and FFA and the greatest plasma glucose, insulin, and IGF-1 was defined as good metabolic status; and the intermediate cluster was defined as average metabolic status. Most dairy cows were classified as having average or good metabolic status, and a limited number of cows had poor metabolic status (10–50 cows per lactation week). On-farm cow data, including dry period length, parity, milk production traits, and body weight, were used to predict good or average metabolic status with 8 machine learning algorithms. Random Forest (error rate ranging from 12.4 to 22.6%) and Support Vector Machine (SVM; error rate ranging from 12.4 to 20.9%) were the top 2 best-performing algorithms to predict metabolic status using on-farm cow data. Random Forest had a higher sensitivity (range: 67.8–82.9% during wk 1 to 7) and negative predictive value (range: 89.5–93.8%) but lower specificity (range: 76.7–88.5%) and positive predictive value (range: 58.1–78.4%) than SVM. In Random Forest, milk yield, fat yield, protein percentage, and lactose yield had important roles in prediction, but their rank of importance differed across lactation weeks. In conclusion, dairy cows could be clustered for metabolic status based on plasma metabolites and metabolic hormones. Moreover, on-farm cow data can predict cows in good or average metabolic status, with Random Forest and SVM performing best of all algorithms.     

Modeling net energy efficiencies as quantitative characteristics in lactating cows

To date, researchers have measured net efficiencies of energy conversion using data from animals in energy chambers. The expense of this approach prevents the establishment of a large data base for quantitative studies. Our purpose was to investigate models that would enable us to use data collectable in normal field conditions to compare dairy cattle for their net energetic efficiency. Data from 357 Holstein cows in seven herds and in various parities consisted of daily measures of DM intake, net energy intake, milk production, biweekly measures of milk components, and bimonthly BW. Eighteen alternative multiple regression models were fitted to each of the cows to estimate simultaneously net efficiency of energy conversion for maintenance, lactation, pregnancy, and BW change during positive energy balance period, negative energy balance period, and whole lactation. Results from several fitted models approximated closely literature results based on data from cows in energy chambers. These comparative results suggest that it is possible to estimate efficiency of energy conversion on individual cows using data obtained from normal animal management situations.     

The use of test day information to predict energy intake of dairy cows in early lactation

This study aimed to validate a previously developed model for the estimation of energy balance in high producing dairy cows from test day information during the first 12 wk of lactation. Monensin (an ionophor) increases the energy status of dairy cows. Gold standard for the validation was a higher energy status, indicated by lower blood ketone body concentrations, lower percent milk fat, and higher milk-yield of monensin-supplemented than control cows in 8 randomized block design feeding trials. Estimated energy intake (eE(intake)) was calculated as estimated energy balance (eEB) plus energy in actual milk produced (in units of MJ(nel)) plus a constant or variable amount of energy required for maintenance. The variable amount was based on BW, while the constant was the average BW in each parity group (1, 2, 3, 4+). Both eEB and eE(intake) were compared between groups of cows with and without monensin supplementation (n = 600 lactations). The trials started with a presupplement period during lactation wk 2 to 5 followed by a supplementation period during lactation wk 6 to 12. During the presupplement period, both eEB and eE(intake) were similar for all cows. At 2, 3, and 8 wk after starting the monensin supplementation, the eEB of the supplemented cows was significantly higher, while eE(intake) was significantly higher throughout the supplementation period. The results were similar for the 2 methods of calculating energy for maintenance, variable or constant. The feed conversion efficiency, calculated as kg of fat-protein corrected milk per MJ(nel) of eE(intake), was highest in first calving cows compared with cows having more lactations, and correlated with standard milk production at trial group level. It was concluded that eE(intake) was a valid measure of net energy absorption.