Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Animal Genomics and Improvement Laboratory » Research » Publications at this Location » Publication #291559

Research Project: Improving Genetic Predictions in Dairy Animals Using Phenotypic and Genomic Information

Location: Animal Genomics and Improvement Laboratory

Title: Dissection of genomic correlation matrices using multivariate factor analysis in dairy and dual-purpose cattle breeds

item Macciotta, N.p. - University Of Sassari
item Dimauro, C - University Of Sassari
item Vicario, D - Collaborator
item Cole, John
item Null, Daniel

Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: 2/23/2013
Publication Date: 7/8/2013
Citation: Macciotta, N.P., Dimauro, C., Vicario, D., Cole, J.B., Null, D.J. 2013. Dissection of genomic correlation matrices using multivariate factor analysis in dairy and dual-purpose cattle breeds. Journal of Dairy Science. 96(E-Suppl. 1):618 (abstr. 541).

Interpretive Summary:

Technical Abstract: SNP effects estimated in genomic selection programs allow for the prediction of direct genomic values (DGV) both at genome-wide and chromosomal level. As a consequence, genome-wide (G_GW) or chromosomal (G_CHR) correlation matrices between genomic predictions for different traits can be calculated. Comparison between G_GEN and G_CHR or between different G_CHR may indicate differences in the genetic control of groups of traits. In this work, a method for comparing genomic correlation matrices based on multivariate factor analysis (MFA) is presented. Two breeds were considered: 3,096 US Holstein and 460 Italian Simmental bulls, with DGV for 31 and 12 productive and functional traits, respectively. Factor analysis was carried out on G_GEN and G_CHR within each breed. In Holstein, between 7 and 9 factors were able to explain 70 to 80% of the original variance, whereas in Simmental on average 3 to 4 latent variables explained about 80% of the variance. In US Holstein, latent factors associated (r >= 0.60) with milk yield traits, milk composition, udder morphology, strength, and functional traits (productive life, SCS, daughter pregnancy rate) were obtained from G_GEN. Differences were observed at the chromosome level. For BTA14, a single factor associated with both milk yield and composition traits was observed. For BTA18, sire calving traits and some conformation traits were associated with the same common factor. In the G_GEN of Italian Simmental, the first latent factor was associated positively with milk yield and milking traits, and negatively with muscularity; the second with daily gain and size; the third to feet and legs and SCS; and the fourth to milk composition traits. On BTA7, one factor is positively associated with daily gain and negatively with milk composition. The MFA was able to detect differences in genetic correlation patterns across the genome, as well as on individual chromosomes, and may be used for preliminary identification of genome regions affecting multiple traits.