plgem.obsStn {plgem}R Documentation

Computation of Observed and Resampled PLGEM-STN statistics

Description

These functions compute observed and resampled signal to noise ratio (STN) values using PLGEM fitting parameters (obtained via a call to function plgem.fit) to detect differential expression in an ExpressionSet ‘data’, containing either microarray or proteomics data.

Usage

  plgem.obsStn(data, plgemFit, covariateNumb=1, baseline.condition=1,
    verbose=FALSE)
  plgem.resampledStn(data, plgemFit, covariateNumb=1, baseline.condition=1,
    iterations="automatic", verbose=FALSE)

Arguments

data an object of class ExpressionSet; see Details for important information on how the phenoData slot of this object will be interpreted by the function.
plgemFit list; the output of ‘plgem.fit’.
covariateNumb integer (or coercible to integer); the covariate used to determine on which samples to fit the PLGEM.
baseline.condition integer (or coercible to integer); the condition to be treated as the baseline.
verbose logical; if TRUE, comments are printed out while running.
iterations number of iterations for the resampling step; if "automatic" it is automatically determined.

Details

The ‘covariateNumb’ covariate (the 1st one by default) in the pData of the ExpressionSet ‘data’ is expected to contain the necessary information about the experimental design. The values of this covariate must be sample labels, that have to be identical for samples to be treated as replicates. In particular, the ExpressionSet ‘data’ must have at least two conditions in the ‘covariateNumb’ covariate; by default the first one is considered the baseline.

PLGEM-STN values are a measure of the degree of differential expression between a condition and the baseline:

PLGEM-STN = [mean(condition)-mean(baseline)] / [modeledSpread(condition)+modeledSpread(baseline)],

where: ln(modeledSpread) = PLGEMslope * ln(mean) + PLGEMintercept

plgem.obsStn determines the observed PLGEM-STN values for each gene or protein in ‘data’. plgem.resampledStn determines the resampled PLGEM STN values for each gene or protein in ‘data’ using a resampling approach; see References for details. The number of iterations should be chosen depending on the number of replicates of the condition used for fitting the model.

Value

plgem.obsStn returns a matrix of observed PLGEM STN values. The rownames of this matrix are identical to the rownames of ‘data’. The colnames represent the different experimental conditions that were compared to the baseline.
plgem.resampledStn returns a list with two items:

RESAMPLED.STN matrix of resampled PLGEM STN values, with rownames identical to those in ‘data’, and colnames representing the different number of replicates found in the different comparisons; see References for details.
REPL.NUMBER the number of replicates found for each experimental condition; see References for details.

Author(s)

Mattia Pelizzola mattia.pelizzola@gmail.com

Norman Pavelka nxp@stowers-institute.org

References

Pavelka N, Pelizzola M, Vizzardelli C, Capozzoli M, Splendiani A, Granucci F, Ricciardi-Castagnoli P. A power law global error model for the identification of differentially expressed genes in microarray data. BMC Bioinformatics. 2004 Dec 17;5:203.; http://www.biomedcentral.com/1471-2105/5/203

Pavelka N, Fournier ML, Swanson SK, Pelizzola M, Ricciardi-Castagnoli P, Florens L, Washburn MP. Statistical similarities between transcriptomics and quantitative shotgun proteomics data. Mol Cell Proteomics. 2007 Nov 19; http://www.mcponline.org/cgi/content/abstract/M700240-MCP200v1

See Also

plgem.fit, plgem.pValue, plgem.deg, run.plgem

Examples

  data(LPSeset)
  LPSfit <- plgem.fit(data=LPSeset)
  LPSobsStn <- plgem.obsStn(data=LPSeset, plgemFit=LPSfit)
  set.seed(123)
  LPSresampledStn <- plgem.resampledStn(data=LPSeset, plgemFit=LPSfit)
  plot(density(LPSresampledStn[["RESAMPLED.STN"]], bw=0.01), col="black", lwd=2,
    xlab="PLGEM STN values",
    main="Distribution of observed and resampled PLGEM STN values")
  lines(density(LPSobsStn, bw=0.01), col="red")
  legend("topright", legend=c("resampled", "observed"), col=c("black", "red"),
    lwd=2:1)

[Package plgem version 1.14.0 Index]