regressionplot {globaltest}R Documentation

Regression plot for Global Test result

Description

Produces a plot which can be used to visualize the effect of specific samples on the test result produced by globaltest.

Usage

regressionplot(testresult, samplenr = NULL)

Arguments

testresult The output of a call to globaltest.
samplenr A vector giving row numbers of samples of interest.

Details

The regressionplot plots, for all pairs of samples, the covariance between the expression patterns against the covariance between their clinical outcomes. Each point in the plot therefore represents a pair of samples. A regression line is fitted through the samples, which visualizes the test result of the function globaltest. A steeply increasing slope indicates a high (possibly significant) value of the test statistic.

An optional argument samplenr can be supplied, giving sample numbers of possibly outlying arrays. In this case, all pairs of arrays involving one of the arrays in samplenr is marked as a red cross, while the other pairs are marked as a blue dot. The blue line which is fitted through all points can now be compared to a red dotted line which is fitted though only the red crosses.

Value

NULL (no output).

Author(s)

Jelle Goeman: j.j.goeman@lumc.nl

References

J. J. Goeman, S.A. van de Geer, F. de Kort and J. C. van Houwelingen, A global test for association of a group of genes with a clinical outcome, Technical Report MI 2003-03, Mathematical Institute, Leiden University. Available from http://www.math.leidenuniv.nl/~jgoeman.

See Also

globaltest, checkerboard, geneplot, permutations.

Examples

if(interactive()){
    data(exampleX)      # Expression data (40 samples; 1000 genes)
    data(exampleY)      # Clinical outcome for the 40 samples
    pathway <- 1:25     # A pathway contains genes 1 to 25
    gt <- globaltest(exampleX, exampleY, test.genes = pathway)
    gt
    regressionplot(gt)
    regressionplot(gt,40)
}

[Package Contents]