agridat (version 1.5)

ilri.sheep: Birth weight and weaning weight of Dorper x Red Maasi lambs

Description

Birth weight and weaning weight of 882 lambs from a partial diallel cross of Dorper and Red Maasi breeds.

Usage

data(ilri.sheep)

source

Case Study 4: Mixed model analysis for the estimation of components of genetic variation in lamb weaning weight. International Livestock Research Institute. http://www.ilri.org/biometrics/CS/case%20study%204/case%20study%204.1.htm Retrieved Dec 2011. Used under Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.

Details

Red Maasai sheep in East Africa are perceived to be resistant to certain parasites. ILRI decided in 1990 to investigate the degree of resistance exhibited by this Red Maasai breed and initiated a study in Kenya. A susceptible breed, the Dorper, was chosen to provide a direct comparison with the Red Maasai. The Dorper is well-adapted to this area and is also larger than the Red Maasai, and this makes these sheep attractive to farmers. Throughout six years from 1991 to 1996 Dorper (D), Red Maasai (R) and Red Maasai x Dorper crossed ewes were mated to Red Maasai and Dorper rams to produce a number of different lamb genotypes. For the purposes of this example, only the following four offspring genotypes are considered (Sire x Dam): D x D, D x R, R x D and R x R. Records are missing in 182 of the lambs, mostly because of earlier death.

References

Baker, RL and Nagda, S. and Rodriguez-Zas, SL and Southey, BR and Audho, JO and Aduda, EO and Thorpe, W. 2003. Resistance and resilience to gastro-intestinal nematode parasites and relationships with productivity of Red Maasai, Dorper and Red Maasai x Dorper crossbred lambs in the sub-humid tropics. Animal Science, 76, 119-136.

Examples

dat <- ilri.sheep
dat <- transform(dat, lamb=factor(lamb), ewe=factor(ewe), ram=factor(ram),
                 year=factor(year))
# dl is linear covariate, same as damage, but truncated to [2,8]
dat <- within(dat, {
  dl <- damage
  dl <- ifelse(dl < 3, 2, dl)
  dl <- ifelse(dl > 7, 8, dl)
  dq <- dl^2
})
  
dat <- subset(dat, !is.na(weanage))

# EDA
require("lattice")
bwplot(weanwt ~ year, dat) # Year effect
bwplot(weanwt ~ factor(dl), dat) # Dam age effect
bwplot(weanwt ~ gen, dat) # Genotype differences
xyplot(weanwt ~ weanage, dat, type=c('p','smooth')) # Age covariate

# case study page 4.18
lm1 <- lm(weanwt ~ year + sex + weanage + dl + dq + ewegen + ramgen, data=dat)
summary(lm1)
anova(lm1)

require("lme4")
lme1 <- lmer(weanwt ~ year + sex + weanage + dl + dq + ewegen + ramgen +
             (1|ewe) + (1|ram), data=dat)
print(lme1, corr=FALSE)
lme2 <- lmer(weanwt ~ year + sex + weanage + dl + dq + ewegen + ramgen +
             (1|ewe), data=dat)
lme3 <- lmer(weanwt ~ year + sex + weanage + dl + dq + ewegen + ramgen +
             (1|ram), data=dat)
anova(lme1, lme2,  lme3)

require("asreml")
# case study page 4.20
m1 <- asreml(weanwt ~ year + sex + weanage + dl + dq + ramgen + ewegen,
             data=dat)
anova(m1)

# case study page 4.26
m2 <- asreml(weanwt ~ year + sex + weanage + dl + dq + ramgen + ewegen,
             random = ~ ram + ewe, data=dat)
anova(m2)

# case study page 4.37, year means
predict(m2, classify="year")$predictions