snm (version 1.20.0)

snm.plot: Display plots for an snm object

Description

Creates a diagnostic plot of the snm fit.

Usage

snm.plot(x, col.by=NULL, ...)
"plot"(x, ...)

Arguments

x

Output from the snm function.

col.by

A factor vector of length equal to the number of arrays providing a grouping by which to color the intensity-dependent effects. Instead of a factor vector, the input may also be a model matrix composed only of 0's and 1's with the number of rows equal to the number of arrays and number of columns less than or equal to the number of arrays.

...

Arguments passed to the plot functions. Not recommended.

Value

Nothing of interest.

Details

A four panel plot composed of the following:

1. Convergence of $pi_0$ estimates over model fitting iterations. The $pi_0$ estimates for each iteration are compared to the $pi_0$ estimate calculated during the final model fit.
2. A scree plot of the principal components analysis of the full model residual matrix.
3. A plot of the estimated intensity-dependent effects.
4. A histogram of the p-values testing each probe for an asssociation with the biological variables (bio.var). All probes to the right of the vertical red line are the least $pi0.hat$ significant probes (i.e., those used in estimating intensity-dependent effects). The dashed horizontal line is the $pi_0$ estimate from the final model fit.

References

Mecham BH, Nelson PS, Storey JD (2010) Supervised normalization of microarrays. Bioinformatics, 26: 1308-1315.

See Also

snm, sim.singleChannel

Examples

## Not run: 
# singleChannel <- sim.singleChannel(12345)
# snm.obj <- snm(singleChannel$raw.data,
# 		      singleChannel$bio.var,
# 		      singleChannel$adj.var,
# 		      singleChannel$int.var, num.iter=10)
# plot(snm.obj, col.by=snm.obj$bio.var) #color by biological group
# plot(snm.obj, col.by=snm.obj$adj.var[,-6]) #color by batch
# ## End(Not run)