- Chen, Y. , Atashi, H. , Wijnrocx, K. , Bérat, H. , Vanlierde, A. , Dehareng, F. , Soyeurt, H. , Gillon, A. & Gengler, N. (2026). Routine Genomic Evaluation of Mid-Infrared-Predicted Methane Emissions in Holstein Cattle: Validation and Implementation of Breeding for Methane Efficiency. Animal, In press:
| Year | 2026 |
|---|---|
| Title | Routine Genomic Evaluation of Mid-Infrared-Predicted Methane Emissions in Holstein Cattle: Validation and Implementation of Breeding for Methane Efficiency |
| Abstract | Enteric methane (CH4) emissions from dairy cattle contribute to greenhouse gas emissions and represent an energy loss for the animal. Genetic selection provides a permanent mitigation strategy, but its implementation requires reliable methane emission phenotypes suitable for routine genetic evaluation. This study describes the final steps toward the development, validation, and operational implementation of a routine genomic evaluation for milk mid-infrared (MIR)-predicted methane emissions (PCH4) in Walloon Holstein. The MIR prediction equation, calibrated using sulfur hexafluoride tracer technique and respiration chamber measurements, achieved a cross-validation R² of 0.68 and a standard error of prediction of 57 g CH₄ d⁻¹. Daily PCH4 records predicted from MIR data, collected between 2007 and 2023 during routine milk sampling, were analysed using a singlestep genomic best linear unbiased prediction random regression test-day model, considering three parity-specific traits for the first three lactations. Heritability estimates for PCH4 ranged from 0.17 to 0.19 across the first three lactations, with very high genetic correlations between lactations (0.87–0.89), supporting the use of the mean genomic estimated breeding value (GEBV) across lactations. Predictions were validated using the linear regression method, regressing GEBV from the full dataset (phenotypes to 2023) on those from the partial dataset (phenotypes to 2019) in young genotyped animals (n = 2 038), yielding a prediction accuracy of 0.83 and dispersion coefficient of 0.90. To enable methane mitigation without compromising current breeding objectives, methane efficiency indices were defined as residual methane after accounting for expected emissions based on production (RMEP), functional (RMEF), or overall economic (RMEG). Approximate genetic correlations were estimated between 37 traits and indices, indicating a slightly unfavourable genetic correlation between RMEP and the functional index (−0.12) and a stronger unfavourable correlation between RMEF and the production index (−0.40). In contrast, RMEG displayed a favourable genetic correlation with the functional index (0.10) and only a slightly unfavourable correlation with the production index (−0.10). Among the evaluated indices, RMEG showed the most balanced pattern of genetic correlations while remaining strongly associated with reduced methane emissions in daughters. Overall, these results demonstrate that milk MIR-predicted methane can be incorporated into routine single-step genomic evaluations, supporting methane mitigation while maintaining current breeding objectives, and the approach can be adopted by Holstein populations in other countries using routine milk MIR and genomic data. |
| Journal | Animal |
| Volume | In press |
| Date | 19/10/2026 |
| Authors | Chen, Y. , Atashi, H. , Wijnrocx, K. , Bérat, H. , Vanlierde, A. , Dehareng, F. , Soyeurt, H. , Gillon, A. & Gengler, N. |