Description of the problem:

Number of classes: 4
Number of genes that passed filtering criteria: 17545
Type of univariate test used: F-test (with random variance model)
Column of the Experiment Descriptors sheet that defines class variable : Scores
WARNING: Distribution assumptions underlying the univariate test random variance model are not satisfied. Random variance model was not used
Multivariate Permutations test was computed based on 20000 random permutations
Permutation p-values for significant genes were computed based on 20000 random permutations

Nominal significance level of each univariate test: 1e-07
Confidence level of false discovery rate assessment: 99 %
Maximum allowed number of false-positive genes: 10
Maximum allowed proportion of false-positive genes: 0.001

Summary of Results:

Number of genes significant at 1e-07 level of the univariate test: 1382
Probability of getting at least 1382 genes significant by chance (at the 1e-07 level) if there are no real differences between the classes: 0

Genes which discriminate among classes:

Table 1 - Sorted by p-value of the univariate test.
The first 1382 genes are significant at the nominal 1e-07 level of the univariate test
With probability of 99 % the first 2622 genes contain no more than 10 false discoveries.
With probability of 99% the first 1926 genes contain no more than 0% of false discoveries.