Object Oriented Microarray Library: Channel Set

The channel set module provides the definition of the complete.channel.set and channel.set classes. See the bottom of the page for an example of how the classes can be used.

Class Name: complete.channel.set

Attributes

data: A list, each of whose entries is a complete.channel object.
name: A character string describing this set of things.

Methods

complete.channel.set(namelist, groupname): This is the only constructor currently provided for these objects. The first argument is a list of the names of individual complete channels. These names are evaluated and the results are strung into a list. The second argument is optional; it defaults to "working set".
extract(object, extractor, ...): This method converts a complete.channel.set into a channel.set by applying the extractor function to every element of the data list. Extra arguments are passed along to the extractor.

Description

An object of the complete.channel.set class represents a collection of related complete channel objects.

Class Name: channel.set

Attributes

data: A list of channel objects.
name: A character string describing this set of things.

Methods

channel.set(namelist, extractor, ep, name): The basic constructor. Only the first argument is required; it must be a list of names of channel objects. The second argument defaults to the value svol.extractor. The third argument contains any additional parameters to be passed along to the extractor; it defaults to zero. The final argument gives a name to the collection; it defaults to "working set".
as.data.frame(object): Convert the object to a data frame by combining the measurements from the list of channels.
as.matrix(object): Convert the object to a matrix by combining the measurements from the list of channels.
summary(object, ...): Write out a summary of the object.
print(object, ...): Print all information about the object.
plot(object, which, eps, name, labels): Produces one or more cluster plots of the object. By default, it produces three different cluster plots, (1) using Euclidean distance, (2) using correlation, (3) converting to binary on/off and using Euclidean distance. By passing a list of numbers in as the optional which argument, you can control which plots are produced. The threshold for the binary conversion is determined by the optional eps argument. If this value is negative, then the minimum value in the data matrix is used, under the assumption that this is a common threshold set earlier. The optional name and labels arguments are used to label the plot; they default to the object's name and the names given to the distinct channels when they were constructed.
transform(object, transformer, parameters): This method applies a transformer to all elements of the data list, returning a new channel.set.
reliable.spots(object, threshold): Returns a logical vector of length equal to the number of genes (rows in each of the channels that are part of this set), where T is interpreted to mean "on in as many samples as possible" and F as "off more than expected".

Description

An object of the channel.set class represents a collection of related samples on which we have made common measurements.

Example

  aa.lister <- read.table('aa_lister.tsv', header=T, row.names=NULL, sep="\t")
  for (i in 1:(dim(aa.lister)[1])) { f.load.clontech(aa.lister[i,]) }
  my.names <- as.vector(aa.lister$varname)
  corrected <- channel.set(my.names)
  above <- apply(as.data.frame(corrected), 2,
                 function(x) { f.above.thresh(x, 0) })
  rely <- reliable.spots(corrected)
  LN.cs <- extract(complete.channel.set(my.names), standard.channel,
	nf=subset.normalize.transform, np=rely, tp=threshold)
  LN <- as.data.frame(LN.cs)