Study On Quality Index And Fatty Acid Characteristics Of Raw Milk From Three Classes Of Dairy Farm

The Chinese dairy industry’s vitalization strategy has been comprehensively promoted,the number of large-scale dairy farms has soared,and the development level is uneven.The raw milk quality indexes and fingerprint data of different dairy farms have not been systematically tested and compared,and the discriminant analysis and evaluation based on stoichiometry have not yet been carried out.A total of 720 samples of raw milk from dairy farms of A,B and C grades in three areas of Golden milk source belt in northern China were collected.Five quality indexes of non-fat milk solids(SNF),Fat(Fat),protein(Pro),acidity(TA)and somatic cell count(SCC)were detected.Fatty acids(FAs)were detected by gas chromatography,and traditional statistical comparisons were made.Principal Component Analysis(PCA)and Orthogonal Partial Least Squares Discriminant Analysis(OPLS-DA)were performed to evaluate the clustering and discriminant ability of quality indexes and fatty acids to dairy farm level,region and season.The results showed that there were significant differences in the quality indexes of raw milk samples from three dairy farms.The index with the greatest difference and the best consistency with dairy farm level was SCC,the total mean(X±1SD,10000/mL)was A(16.4±4.08),B(22.5±6.03),C(29.6±9.83).There were significant differences in Fat in both total sample and regional grade 3 dairy farms,but it lacked consistency with cattle farm level.The total mean was(X± 1SD,g/100g),Grade A(3.88±0.20),Grade B(3.94±023)and grade C(3.90±0.25).Pro,SNF and TA showed statistical differences among regional three cattle farms,and were not consistent with the cattle farm level.The Pro mean value was less than 3.2g/100g in all three grade ranches.The quality indexes of Chifeng were statistically different among regions.Pro,Fat and SNF were higher than those of Shuozhou and Bayannur.Shuozhou had the highest SCC.TA of Bayannur was higher than the other two regions.The quality indexes showed seasonal differences.Pro,Fat,SNF and TA in winter and spring were higher than those in summer and autumn,but the absolute and relative differences were not obvious.The SCC of grade A and Grade B dairy farms in summer and autumn was significantly higher than that in winter and spring,but there was no significant difference in grade C and total samples,which may be related to the higher SCC and large variation in different seasons.In the PCA and OPLS-DA analysis of quality indicators,there was a trend of clustering separation among the total samples of A,B and C cattle farms,and there was obvious clustering separation among the three cattle farms in the three places.The clustering distance of raw milk samples in the whole A and B pastures was close,while that in the C pasture was far away and the most discrete.OPLS-DA had a high accuracy rate of 86%～95%in the discrimination of the three-level cattle farms in the same area,but only 58.8%in the discrimination of the total sample cattle farms,which also showed that the consistency of the three-level cattle farms in the three places was not high.The index that contributed the most to the classification was SCC,and the C-class dairy farm was scattered because of the large variation of SCC.Fat,Pro and SNF in the three regions were alternately the factors with high classification contribution rate,but they were all inconsistent with the cattle farm level.TA was the lowest influence on classification.FAs fingerprint clustering analysis showed that the ability to distinguish dairy farm level,region and season was better than the quality index.The correct rate of OPLS-DA cross-validation for cattle farm level was 76.7%,and the correct rate of OPLS-DA cross-validation for cattle farm level was 90.9%～94.5%.The accuracy of regional discrimination was 94.1%～97.8%better than that of pasture level.The correct discriminant rate of winter,spring and summer was 99.4%.In conclusion,FAs was sensitive to breeding level,regional and seasonal differences.PUFA of Grade A dairy farm was significantly higher than that of grade B and Grade C dairy farm.There was no difference in SFA of dairy farms in the three regions,and PUFA,MUFA and TFA in Bayannur were higher than those in Chifeng and Shuozhou.It was of great reference,promotion and significance to establish the database of raw milk quality and fingerprint,to carry out more scientific classification,data-based,digital scientific management and scientific feeding of dairy farms,which was worthy of in-depth research and application.