机构地区:[1]内蒙古农业大学食品科学与工程学院,内蒙古呼和浩特010018
出 处:《食品研究与开发》2022年第10期172-180,共9页Food Research and Development
基 金:国家自然科学基金地区科学基金项目(31760489)。
摘 要:采集国内3家乳品企业荷斯坦牛有机和非有机的原乳与超高温灭菌(ultra-high temperature,UHT)乳,采用气相色谱法测定其脂肪酸(fatty acids,FAs)构成(称为FAs指纹),并进行主成分分析(principal component analysis,PCA)和描述性统计比较,建立软独立建模分类(soft independent modeling of class analogies,SIMCA)判别模型。PCA分析发现原乳和UHT乳均按有机和非有机饲养模式分开聚类,说明有机与非有机乳的FAs指纹不同;但不同企业有机与非有机牧场原乳聚类位置不一致,说明不同企业奶牛饲养模式及标准都有较大差异;不同企业有机和非有机UHT乳聚类的位置和趋势有一致性,这可能与UHT乳生产使用大罐多牧场混合原奶及标准化工艺有关。有机和非有机乳间脂肪酸独立样本t检验多数有显著或极显著差异,但整体FAs均值配对t检验差异不显著,即传统统计不能区分有机和非有机乳FAs复杂的整体差异。总体上有机乳中α-亚麻酸(C18∶3n3)和亚油酸(C18∶2n6c)含量显著高于非有机乳,非有机乳C16∶0和C16∶1含量高;3家乳企有机和非有机乳之间关键FAs的比值不同,再次说明不同企业奶牛饲养模式或饲料不同。有机和非有机UHT乳SIMCA判别模型内部和外部验证正确率分别为100%和90.5%,FAs指纹建模判别有机与非有机UHT乳可行。Here organic and non-organic Holstein raw milk and ultra-high temperature(UHT)milk from three Chinese dairy companies were collected,and gas chromatography was used to detect the composition of fatty acids(FAs).Principal component analysis(PCA)and descriptive statistics were conducted,and soft independent modeling of class analogies(SIMCA)was established.As a result,PCA showed that raw milk and UHT milk were both clustered according to the organic and non-organic feeding patterns,indicating that FAs fingerprints between organic and non-organic milk were different.However,the clustering positions of organic and non-organic raw milk from different dairy companies were not consistent,implying that the feeding pattern and feeding standards of different dairy farms were distinct;while the clustering positions and trend of the organic and non-organic UHT milk from different dairy companies were consistent,which may be contributed to the mixture of multiple farm's raw milk in ultra big bulk tank and the standardized procedure during UHT milk production.The independent samples t-test showed that most single fatty acid had significant or extremely significant difference between organic and non-organic milk,but there was no remarkable difference in overall FAs mean in the paired t-test;in other words,traditional statistics could not distinguish the complex overall differences of FAs in organic and nonorganic milk.Generally,the content ofα-linolenic acid(C18∶3n3)and linoleic acid(C18∶2n6c)in organic milk was markedly higher than that in non-organic milk,while the content of C16∶0 and C16∶1 was higher in non-organic milk.The ratio of key FAs between organic and non-organic milk in the three dairy companies were differ-ent,which further suggested that feeding pattern or feed formula varied among different dairy companies.The ac-curacy rates of SIMCA internal and external validation for organic and non-organic UHT milk authentication were 100%and 90.5%,respectively,and FAs fingerprint modeling was feasible to distingu
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...