Lipidomics revealed the geographical origin and breed variability of sheep milk

Breed and geographical origin are critical factors influencing the composition of dairy products. This study employed untargeted lipidomics to investigate the milk lipid composition of three groups of sheep: DairyMeade (China, n = 11), East Friesian (China, n = 20) and East Friesian (the Netherlands, n = 15), with all milk samples collected in September. The objective was to assess the effects of breed (DairyMeade vs. East Friesian, both from China) and geographical origin (East Friesian from China vs. the Netherlands) on milk lipid profiles. A total of 17 lipid classes and 789 lipid molecules were identified from the three groups using the SIRIUS software. Combined with the criteria of p-value <0.001, VIP (Variable Importance in the Projection) > 1, and fold change >2 or < 0.5, 30 and 40 lipid molecules could serve as potential indicators for identifying sheep breeds and geographical origins, respectively. Furthermore, linear discriminant analysis was applied to construct classification models based on statistically significant lipids. For breed and origin discrimination, four and six variables were selected via stepwise regression to establish Fisher and Bayes discriminant functions, respectively. Both models achieved 100 % overall classification accuracy and 100 % accuracy under leave-one-out cross-validation. This study offers comprehensive insights into the lipid composition of sheep milk and establishes a robust scientific basis for the authentication and traceability of sheep milk products based on breed and geographical origin.

Text 124231.pdf - Published Version Restricted to Repository staff only Download (5MB) Official URL or Download Paper: https://www.sciencedirect.com/science/article/pii/...

Actions (login required)

View Item