RankingFoxDimmic {GeneSelector} | R Documentation |
Performs a two-sample Bayesian t test on a gene expression matrix using the method of Fox and Dimmic (2006).
RankingFoxDimmic(x, y, type = "unpaired", m = 4, pvalues = TRUE, gene.names = NULL, ...)
x |
A |
y |
If |
type |
|
m |
The number of similarly expressed genes to use for calculating Bayesian variance and prior degrees of freedom. The default value suggested by Fox and Dimmic is currently 4, s. note. |
pvalues |
Should p-values be computed ? Default is |
gene.names |
An optional vector of gene names. |
... |
Currently unused argument. |
An object of class GeneRanking.
Although the test of Fox and Dimmic is very similar to that of Baldi and Long; there are various slight differences, in particular with respect to the computation of the Bayesian variance.
Martin Slawski
Anne-Laure Boulesteix
Fox, R.J., Dimmic, M.W. (2006).
A two sample Bayesian t-test for microarray data.
BMC Bioinformatics, 7:126
RepeatRanking, RankingTstat, RankingFC, RankingWelchT, RankingWilcoxon, RankingBaldiLong, RankingLimma, RankingEbam, RankingWilcEbam, RankingSam, RankingShrinkageT, RankingSoftthresholdT, RankingPermutation
## Load toy gene expression data data(toydata) ### class labels yy <- toydata[1,] ### gene expression xx <- toydata[-1,] ### run RankingFoxDimmic FoxDimmic <- RankingFoxDimmic(xx, yy, type="unpaired")