pvalues.manova: Computes pvalues for multi-class differential expression
In Pigengene: Infers biological signatures from gene expression data

Description Usage Arguments Value Note Author(s) References See Also Examples

Passes the arguments to manova, which performs multi-class analysis of variance.

1	pvalues.manova(Data, Labels)

`Data`	A matrix or data frame containing the (expression) data, with genes corresponding to columns and rows corresponding to samples. Rows and columns must be named.
`Labels`	A (preferably named) vector containing the Labels (condition types). Names must agree with rows of `Data`

A list with following elements:

`call`	The call that created the results
`pvals`	The matrix of pvalues with columns "pValue", "FDR", "Bonferroni". Rows are named according to genes, the columns of `Data`.
`manovaFit`	The full output of `manova` function.

oneway.test function is a better generalizatoion to Welch's t-tst from 2-calsses to multi-class because it dose not assume that the variaces are necessarly equal. However, in practice, with "enough number of samples", the two approaches will lead to similar p-values.

Amir Foroushani

Krzanowski, W. J. (1988) _Principles of Multivariate Analysis. A User's Perspective._ Oxford.

Hand, D. J. and Taylor, C. C. (1987) _Multivariate Analysis of Variance and Repeated Measures._ Chapman and Hall.

B. L. Welch (1951), On the comparison of several mean values: an alternative approach.

oneway.test, manova, compute.pigengene

     data(eigengenes33)
     d1 <- rbind(eigengenes33$aml,eigengenes33$mds)
     Labels <- c(rep("AML",nrow(eigengenes33$aml)),rep("MDS",nrow(eigengenes33$mds)))
     names(Labels) <- rownames(d1)
     ps <- pvalues.manova(Data=d1, Labels=Labels)
     plot(log10(ps$pvals[,"Bonferroni"]))