Dear R-helper:
I measured 40 different mouse strains. In 10 of them, I measured 2 males and
4 females for each strain. In another 10 of them, I measured 4 males and 2
females for each strain. In the remaining 20, I measured 3 males and 3 females
for each strain. Totally, I have 240 data for 40 strains each of them have 6
data.
The model for the data can be written as, pheno=mu+sex+strain+e, where mu
(mean) and sex (two levels) are fixed effects, and strain (40 levels) and e are
random effects:
result1 <- lme(pheno~sex, random=list(strain=~1),data=dat) (model_1)
This works and I obtained variances for both strain and residual. However,
these 40 strains are highly correlated. I don't know to how to incorporate
dependence among these strains into the above model. I already obtained a
correlation matrix for the 40 strains. The below R does not work:
result2 <- lme(pheno~sex, random=list(strain=~1),
correlation=corSymm(value=c(0.8,0.7,...,0.4),form =~ Strain), data=dat)
(model_2)
I will greatly appreciate if you can give some suggestions
Pengyuan Liu
Dept of Surgery
Washington Univ in St Louis
---------------------------------
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
