|Van vleck, Lloyd|
Submitted to: Journal of Animal Science
Publication Type: Peer reviewed journal
Publication Acceptance Date: 3/31/1999
Publication Date: N/A
Citation: Interpretive Summary: Previously animal breeders have been forced to assume that the same genes influence performance at different ages. Statistical techniques and computing power now allow for testing that assumption. Fleece measurements of ewes of Columbia, Polypay, Rambouillet, and Targhee breeds were divided into measurements taken when ewes were young (Y, 1 yr), middle aged (M, 2-3 3yr) and older aged (O, >3 yr). Correlations between expressions of genotyp at different ages were estimated. If the correlation is greater than .80, the usual conclusion is that expression is similar enough that the assumption of same genes affecting performance is adequate. For staple length, correlation between expression in young and middle aged ewes ranged from .69 to .94 (average of .82) indicating that distinction between ages may not be important. For fleece weight, correlations between genetic expressions at different ages were generally large (.61 to .95, for young with middle ages; .72 to .95, for young with older ages; and .94 to .98, for middle and older ages). The correlations suggest little advantage to treating expressions at different ages as separate, correlated traits. For fleece grade, genetic correlations were more variable (.34 to .96, for young with middle ages; -.00 to .89 for young with older ages; and .66 to .98, for middle with older ages). For fleece grade, there may be some advantage to considering measurements at one year of age as a different trait from measurements at later ages. Heritabilities and phenotypic variation were similar for all age classes which suggests only correlations need be considered in deciding whether measurements at different ages should be analyzed as the same or different traits.
Technical Abstract: Genetic parameters for wool traits for Columbia (C), Polypay (P), Ramboui- let (R), and Targhee (T) breeds of sheep were estimated with single and multiple-trait analyses by REML with animal models. Traits considered were fleece grade (FG), fleece weight (FW) and staple length (SL). Total observations ranged from 11,673 to 34,746 for FG and FW, and from 3,590 to 11,641 for SL for the four breeds. For single-trait analyses, data were divided by age of ewe: young ages (Y, age of 1 yr), middle ages (M, ages of 2 and 3 yr), and old ages (0, age greater than 3 yr). Average heritability estimates for FG decreased from .42 at 1 year to .37 for older ages. For FW heritability estimates averaged .52, .57, and .55 with increasing age. Heritability estimates for SL averaged .54 for Y and M. Few older ewes had SL measurements. After single-trait analyses, new data sets were created for three-trait analyses with traits defined by three age classes when animals were measured. Heritability estimates from three-trait analyses were somewhat greater than those from single-trait analyses. For FG, the genetic correlations averaged .72 for Y with M, .42 for Y with O and .86 for M with O. For FW the average genetic correlations were greater: .81, .83, and .98. For SL the average genetic correlation for Y with M was .82. Estimates of genetic correlations across ages varied considerably among breeds. Average estimates of correlations suggest that for FG, the trait may need to be defined by age. For FW and SL, however, the average correlations suggest no need to define those traits by age.