Processor-class {PreProcess}

Class "Processor"

Description

A Processor represents a function that acts on the data of a some object to process it in some way. The result is always another related object, which should record some history about exactly how it was processed.

Usage

## S4 method for signature 'Channel,Processor'
process(object, action, parameter=NULL)
## S4 method for signature 'Processor'
summary(object, ...)

Arguments

object

In the process method, a Channel object. In the summary method, a Processor object

action

A Processor object used to process a Channel.

parameter

Any object that makes sense as a parameter to the function represented by the Processor action

...

Additional arguments are as in the underlying generic methods.

Value

The return value of the generic function process is always an object related to its Channel input, which keeps a record of its history. The precise class of the result depends on the function used to create the Processor.

Slots

f:

A function that will be used to process microarray-related object

default:

The default value of the parameters to the function f

name:

A string containing the name of the object

description:

A string containing a longer description of the object

Methods

process(object, action, parameter)

Apply the function represented by action to the Channel object, updating the history appropriately. If the parameter is NULL, then use the default value.

summary(object, ...)

Write out a summary of the object.

Pre-defined Processors

The library comes with several Processor objects already defined; each one takes a Channel as input and produces a modified Channel as output.

PROC.SUBTRACTOR

Subtracts a global constant (default: 0) from the data vector in the Channel.

PROC.THRESHOLD

Truncates the data vector below, replacing the values below a threshold (default: 0) with the threshold value.

PROC.GLOBAL.NORMALIZATION

Normalizes the data vector in the Channel by dividing by a global constant. If the parameter takes on its default value of 0, then divide by the 75th percentile.

PROC.LOG.TRANSFORM

Performs a log transformation of the data vector. The parameter specifies the base of the logarithm (default: 2).

PROC.MEDIAN.EXPRESSED.NORMALIZATION

Normalizes the data vector by dividing by the median of the expressed genes, where “expressed” is taken to mean “greater than zero”.

PROC.SUBSET.NORMALIZATION

Normalizes the data vector by dividing by the median of a subset of genes. When the parameter has a default value of 0, then this method uses the global median. Otherwise, the parameter should be set to a logical or numerical vector that selects the subset of genes to be used for normalization.

PROC.SUBSET.MEAN.NORMALIZATION

Normalizes the data vector by dividing by the mean of a subset of genes. When the parameter has a default value of 0, then this method uses the global mean. Otherwise, the parameter should be set to a logical or numerical vector that selects the subset of genes to be used for normalization.

Author(s)

Kevin R. Coombes [e-mail]

See Also

Channel, CompleteChannel, process, Pipeline

Examples

showClass("Processor")

## simulate a moderately realistic looking microarray
nc <- 100
nr <- 100
v <- rexp(nc*nr, 1/1000)
b <- rnorm(nc*nr, 80, 10)
s <- sapply(v-b, max, 1)
ct <- ChannelType('user', 'random', nc, nr,  'fake')
subbed <- Channel(name='fraud', parent='', type=ct, vec=s)
rm(ct, nc, nr, v, b, s)		# clean some stuff

## example of standard data processing
nor <- process(subbed, PROC.GLOBAL.NORMALIZATION)
thr <- process(nor, PROC.THRESHOLD, 25)
processed <- process(thr, PROC.LOG.TRANSFORM, 2)

summary(processed)

par(mfrow=c(2,1))
plot(processed)
hist(processed)

par(mfrow=c(1,1))
image(processed)

rm(nor, thr, subbed, processed)