The interpretation of mutual relationship among milk fatty acids (FA) is not easy due to the high number of FA contained in milk fat and to the high degree of correlation among them. Multivariate analysis includes different statistical approaches that could help explaining complex pattern of variables. In this study, Multivariate Factor Analysis (MFA) was used to decompose the correlation matrix of 47 FA and milk production traits (milk yield and protein and fat content) measured in 300 Italian Holstein Friesian cows reared in the North of Italy in 23 commercial dairy farms, representative of the intensive dairy system. MFA was able to extract seven latent factors with specific biologic meaning: secretion of Long Chain FA (KLCFA), mammary FA de novo synthesis (Km), rumen biohydrogenation (Kbh), synthesis of odd chain FA (Ko), synthesis of branched chain FA (Kb), mammary desaturation activity (Kd), milk yield (Kmy). According to the pattern of communalities of the factor analysis, C18:3c9c12c15 was the only FA, along with C18:2t11c15, to be uncorrelated with the other variables and it seemed to be excluded by the metabolic pattern described by the seven factors. The desaturation products of the SCD enzyme were independently associated to three latent factors, suggesting new insights in the regulation of SCD activity. Factors were considered as new quantitative phenotypes related to prominent features of milk FA profile. With the aim of evaluating the feeding regimen and animal effects, latent factors were analysed with a mixed model, which considered the fixed effect of lactation stage, parity, some feeding regimen characteristics and the random effect of bull. Lactation stage significantly affected Km and Kmy factors. In perspective, the seven factors extracted by applying MFA analysis to milk FA composition could be considered as new and more informative traits to test the effect of endogenous and exogenous variation factors.

Investigating mutual relationship among milk fatty acids by multivariate factor analysis in dairy cows

CONTE, GIUSEPPE;SERRA, ANDREA;CAPPUCCI, ALICE;MELE, MARCELLO
2016-01-01

Abstract

The interpretation of mutual relationship among milk fatty acids (FA) is not easy due to the high number of FA contained in milk fat and to the high degree of correlation among them. Multivariate analysis includes different statistical approaches that could help explaining complex pattern of variables. In this study, Multivariate Factor Analysis (MFA) was used to decompose the correlation matrix of 47 FA and milk production traits (milk yield and protein and fat content) measured in 300 Italian Holstein Friesian cows reared in the North of Italy in 23 commercial dairy farms, representative of the intensive dairy system. MFA was able to extract seven latent factors with specific biologic meaning: secretion of Long Chain FA (KLCFA), mammary FA de novo synthesis (Km), rumen biohydrogenation (Kbh), synthesis of odd chain FA (Ko), synthesis of branched chain FA (Kb), mammary desaturation activity (Kd), milk yield (Kmy). According to the pattern of communalities of the factor analysis, C18:3c9c12c15 was the only FA, along with C18:2t11c15, to be uncorrelated with the other variables and it seemed to be excluded by the metabolic pattern described by the seven factors. The desaturation products of the SCD enzyme were independently associated to three latent factors, suggesting new insights in the regulation of SCD activity. Factors were considered as new quantitative phenotypes related to prominent features of milk FA profile. With the aim of evaluating the feeding regimen and animal effects, latent factors were analysed with a mixed model, which considered the fixed effect of lactation stage, parity, some feeding regimen characteristics and the random effect of bull. Lactation stage significantly affected Km and Kmy factors. In perspective, the seven factors extracted by applying MFA analysis to milk FA composition could be considered as new and more informative traits to test the effect of endogenous and exogenous variation factors.
2016
Conte, Giuseppe; Serra, Andrea; Cremonesi, P.; Chessa, S.; Castiglioni, B.; Cappucci, Alice; Bulleri, E.; Mele, Marcello
File in questo prodotto:
File Dimensione Formato  
Conte et al., 2016.pdf

solo utenti autorizzati

Descrizione: articolo principale
Tipologia: Versione finale editoriale
Licenza: NON PUBBLICO - Accesso privato/ristretto
Dimensione 654.21 kB
Formato Adobe PDF
654.21 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/801349
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 32
  • ???jsp.display-item.citation.isi??? 31
social impact