Description Usage Arguments Author(s) Examples
This function produces plots useful for diagnosing problems with an assumed study-specific model.
1 | snm.diagnostic.plot(obs.fit, obs.stat, snm.obj, basisSplineFunction, n.iter)
|
obs.fit |
List of estimated coefficient matrices and residuals from full and reduced models |
obs.stat |
List of statistic obtained from model fit. |
snm.obj |
An object of class snm. |
basisSplineFunction |
Basis spline functions for intenisty-dependent functions |
n.iter |
Current iteration number |
Brig Mecham <brig.mecham@sagebase.org>
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | ##---- Should be DIRECTLY executable !! ----
##-- ==> Define data, use random,
##-- or do help(data=index) for the standard data sets.
## The function is currently defined as
function (obs.fit,obs.stat,snm.obj,
basisSplineFunction)
{
par(mfrow=c(2,2))
bds <- ifelse(snm.obj$bin.densities > 500, 500, snm.obj$bin.densities)
plot(rowMeans(snm.obj$M.pooled),
bds,ylim=c(0,500),
xlab="Estimated RNA Concentration", ylab="Probes per bin", main="Null probes per bin")
u <- fast.svd(obs.fit$res1,tol=0);
plot(u$d^2 / sum(u$d^2), ylim=c(0,1),
main="Latent Structure",
xlab="Surrogate Variable",
ylab="Percent Variance Explained")
if(dim(snm.obj$dat)[1] > 5000) {
th <- sample(dim(snm.obj$dat)[1], 5000)
}else{
th <- 1:dim(snm.obj$dat)[1]
}
oo <- order(snm.obj$M[th,1])
plot(snm.obj$M[th[oo],1], snm.obj$array.fx[th[oo],1],
type="l",lwd=3, ylim=range(snm.obj$array.fx[th[oo],]),
xlab="Estimated RNA Concentration", ylab="Array Effect",main="Estimated Array Effects")
sapply(2:dim(snm.obj$r.dat)[2],function(id) {
oo <- order(snm.obj$M[th,id])
points(snm.obj$M[th[oo],id], snm.obj$array.fx[th[oo],id],type="l",lwd=3)
}) -> hmm
hist(obs.stat$pval, xlab="P values",main="P value distribution");
abline(v=sort(obs.stat$pval)[length(obs.stat$pval)*(1-snm.obj$pi0)],col="red")
}
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.