lvs.fit {FLUSH.LVS.bundle} | R Documentation |
Selects the Least Variant Set of genes, accordingly to the chosen proportion of genes expected not to vary between arrays.
lvs.fit(object, proportion = 0.6, DF=10,...)
object |
an object of class RA or a matrix with residual standard errors and array effects. |
proportion |
the proportion below which genes are expected not to vary between samples. Default is set to 0.6. |
DF |
the degrees of freedom used for the bi-spline in the quantile regression. |
... |
... |
a TRUE/FALSE vector with the seleted ID probes from the quantile regression.
Stefano Calza <calza@med.unibs.it>, Davide Valentini and Yudi Pawitan.
S. Calza et al. 'Normalization of oligonucleotide arrays based on the least variant set of genes' (2008, BMCBioinformatics).
lvs
, normalize.lvs
,
compute.RA
,RA-class
,
normalize.AffyBatch.lvs
## Not run: # Starting from an Affibatch object called aBatch data.RA <- compute.RA(aBatch) lvs.id <- lvs.fit(data.RA, proportion=0.6) lvs.prep <- expresso(aBatch, normalize=FALSE, bgcorrect.method="mas",pmcorrect.method="mas", summary.method="mas") normalize.AffyBatch.lvs(lvs.prep,lvs.id=lvs.id) ## End(Not run)