Skip to main content
Dairy cattle

Towards a single-step genomic evaluation of cheese-making traits predicted from MIR spectra in eight French dairy cattle breeds

Authors
  • Anais Maupomé (Université Paris-Saclay)
  • Delphine Duclos (French Livestock Institute (IDELE))
  • Marine Gelé (French Livestock Institute (IDELE))
  • Olivier Bulot (BGS)
  • Aurélia Baur (Eliance)
  • Nicolas Gaudillière (Eliance)
  • Valérie Wolf (CEL 25-90)
  • Didier Boichard (Université Paris-Saclay)
  • Sébastien Fritz (Eliance)
  • Marie-Pierre Sanchez (Université Paris-Saclay)

Abstract

The FROM'MIR project (2015-2019) developed equations predicting cheese-making properties (CMP) from mid-infrared (MIR) spectra in the Montbéliarde breed, enabling the first genomic evaluation for these traits. The FROM4ALL project extends these predictive models and genomic tools to eight additional French dairy breeds (Abondance, Jersey, Brown Swiss, Tarentaise, Simmental, Normande, Vosgienne, and Holstein). Seven CMP were predicted from MIR spectra: fresh curd yield (CYfresh), curd yield in dry matter (CYDM), curd yield in protein and fat (CYFAT-PROT), ratio of curd firming time to coagulation time for soft cheese (K10/RTCSC), curd firmness at RCT for soft cheese (aSC), curd firmness at RCT for pressed cooked cheese (aPCC), and curd firmness at 2à—RCT for soft cheese (a2SC). The objectives of this study were to estimate genetic parameters for these traits and to implement single-step genomic evaluations. After quality control, the dataset comprised more than 50 million test-day records across the eight breeds. For genetic parameter estimation, 15,548,851 records from 1,950,417 primiparous cows in 27,485 herds were retained, with a minimum of three test-day records per cow. Genetic parameters were estimated using a repeatability animal model implemented in Wombat. Heritability estimates were low to moderate and consistent across breeds, aligning with previous results in Montbéliarde. Estimates ranged from 0.14 to 0.63 for cheese yields and from 0.28 to 0.49 for coagulation traits. Genetic correlations displayed clear and biologically coherent patterns: coagulation traits were strongly positively correlated (0.90-1.00), except for K10/RTCSC. As expected, faster coagulation (lower K10/RTCSC) was associated with firmer curd (aSC, a2SC, aPCC). K10/RTCSC was negatively correlated with all traits (−0.62 to −0.86) except CYFAT-PROT (0.46-0.71), indicating that faster coagulation improves curd firmness and fresh/dry matter yields. CMP exhibited moderate negative correlations with milk yield but generally strong positive correlations with milk protein and fat contents. Associations with somatic cell count were weak and occasionally inconsistent across breeds. Single-step genomic evaluations for CMP provide accurate results and slightly positive genetic trends were observed, in agreement with the trends for milk contents. Overall, these results confirm that CMP are heritable across a wide range of French dairy breeds and provide a robust foundation for improving milk processing quality through genetic selection. The FROM4ALL project was funded by APIS-GENE.

Keywords: 2026

How to Cite:

Maupomé, A., Duclos, D., Gelé, M., Bulot, O., Baur, A., Gaudillière, N., Wolf, V., Boichard, D., Fritz, S. & Sanchez, M., (2026) “Towards a single-step genomic evaluation of cheese-making traits predicted from MIR spectra in eight French dairy cattle breeds”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286612. doi: https://doi.org/10.31274/wcgalp.24066

Rights: 1

Downloads:
Download PDF
View PDF

63 Views

18 Downloads

Published on
2026-02-25

Peer Reviewed