medianPolishIncremental: Median polish fit

In steveneschrich/IncrementalRMA: Affymetrix-based RMA Processing Using Fixed Parameters

Description

Given an existing median polish model, apply to new data and return resulting median polish.

Usage

medianPolishIncremental(pms, probe_names, params = NULL)

Arguments

pms	Perfect-match matrix representing the gene expression.
probe_names	Array of probenames such as returned from probeNames-methods
params	An incremental parameter list as generated by parameterizeRMA

Details

The RMA approach involves using median polish to summarize a set of probes for a set of samples. Briefly, median polish removes row medians then column medians repeatedly until a stopping criteria is achieved. See medpolish for details on the median polish.

The incremental nature of this median polish approach means that final state probe-wise effects have already been calculated (and stored). Thus for a sample, the precalculated probe-wise effects are subtracted from the signal and a probeset-wide median is calculated. This is the used as the probeset expression estimate.

Value

Expression values corresponding to probesets defined by the probe_names parameter.

Note

The incremental median polish information is stored within the params object as the list probeEffects.

If the params object is NULL then a default median polish is performed.

Examples

## Not run: 
medianPolishIncremental(pm(AffyBatch), probeNames(AffyBatch), NULL)

## End(Not run)

steveneschrich/IncrementalRMA documentation built on Dec. 23, 2021, 5:32 a.m.