Contribution of milk mid-infrared spectrum to improve the accuracy of test-day body weight predicted from stage, lactation number, month of test and milk yield


  • Soyeurt, H. , Froidmont, E. , Dufrasne, I. , Hailemariam, D. , Wang, Z. , Bertozzi, C. , Colinet, F. , Dehareng, F. & Gengler, N. (2019). Contribution of milk mid-infrared spectrum to improve the accuracy of test-day body weight predicted from stage, lactation number, month of test and milk yield. Livestock Science, 227: 82-89.
Type Journal Article
Year 2019
Title Contribution of milk mid-infrared spectrum to improve the accuracy of test-day body weight predicted from stage, lactation number, month of test and milk yield
Journal Livestock Science
Volume 227
Pages 82-89
Abstract A regular and repeated recording of body weight (BW) is useful information for herd management. BW can be predicted regularly from animal characteristics such as age, lactation number, or lactation stage. Those traits are unfortunately animal unspecific. Adding animal specific information, which can be easily obtained on a large scale, to the BW prediction would be of utmost importance. There are good scientific reasons to suspect links between BW and animal specific characteristics, available in a repeated fashion, such as milk yield and milk composition. This study aimed to demonstrate the feasibility of predicting test-day BW from stage, lactation number, month of test, milk yield and mid-infrared spectra, representing milk composition. Five models were tested initially from 721 BW records collected in 6 herds: day in milk?+?number of lactation (equation 1a); equation1a?+?milk yield (equation 1b); only spectral data (equation 1c); equation 1c?+?equation 1a (equation 2); equation 2?+?milk yield (equation 3). Then 3 other equations included the same explicative variables, except that the spectral data were regressed using second order Legendre Polynomials (PL) to take into account changes of spectral data within lactation. Equation 1a and 1b were built using linear regressions and equation 1c until 3 were built using partial least square regressions. These 3 last equations had a higher number of factors. Adding of MIR data in the equation increased of 7% the values of cross-validation R² (R²cv). Potential BW outliers were discarded using a residual analysis based on equation 3. From 662 records, the following statistical parameters were obtained: the calibration coefficient of determination (R²c) = 0.65, R²cv?=?0.61, calibration root mean squared error of prediction (RMSEP)=38?kg, and RMSEPcv=40?kg. Low variation of R²c and RMSEPc values obtained from the herd validation confirmed the herd independence of predictions. However, large variability was observed for RMSEPv (37 to 64?kg) suggesting the need to increase the dataset in order to improve the robustness of the equation. By applying the equations on a large spectral database, it was confirmed that the addition of MIR data allows to better model the BW evolution within lactation. Based on these preliminary results, and if a larger validation confirms these findings, this approach could be used to develop equations that are better able to assess BW throughout lactation(s), BW being an important element for management and selection tools.
File
Authors Soyeurt, H. , Froidmont, E. , Dufrasne, I. , Hailemariam, D. , Wang, Z. , Bertozzi, C. , Colinet, F. , Dehareng, F. & Gengler, N.