condPlot (modules, number, eset, col = "white", all = TRUE, sep = NULL, sepcol = "grey", val = TRUE, srt = 90, adj.above = c(0, 0.5), adj.below = c(1, 0.5), plot.only = seq_len(ncol(eset)), ...)
ISAModules
object.ExpressionSet
or ISAExpressionSet
object. This is needed for calculating the scores of the samples
that are not in the module, see the all
argument. If an
ExpressionSet
object is supplied, then it is normalised by
calling ISANormalize
on it.TRUE
, the default), or just the ones that are included in the
module.NULL
or a numeric vector. If not NULL
, then
the bars are separated at the given positions with vertical
lines. This is useful if you want to subdivide the samples into
groups.sep
argument), if they are plotted.text
function.text
, see its manual
for details.text
, see its manual
for details.barplot
.condPlot
creates a barplot for the sample scores of an ISA
transcription module. Each sample is represented as a bar.These plots are useful if you want to show that a given transcription module separates the samples into two (or more) groups. You can assign different colors to the samples, based on some external information, e.g. case and control samples can be colored differently. In most cases the scores are between minus one and one, but this is not necessarily true.
It is possible to assign scores to samples that are not part of the
module, but this requires performing one (more precisely half) ISA
iteration step. Currently the function always performs this extra
step, even if the out-of-module samples are not plotted. Because the
sample scores in a module are only approximately constant during ISA
iteration, it might be possible that the plotted scores are slightly
different than the ones stored in the modules
variable.
ISA
and ISAModules
.data(ALLModulesSmall)
library(ALL)
data(ALL)
col <- ifelse(grepl("^B", ALL$BT), "darkolivegreen", "orange")
condPlot(ALLModulesSmall, 1, ALL, col=col)
Run the code above in your browser using DataLab