ExampleDatasets

Artificial data generated from the F distribution

Permutation values are obtained by randomly drawing samples from the F distribution with parameters (degrees of freedom) 5 and 10. This dataset contains 1000 permutation values for each of five events/experiments. The correct theoretical permutation test P-values for these five events are found when evaluating the cdf of the F distribution at the values of the original statistic (on the first row in the dataset). These P-values are: 10-1, 10-2, 10-3, 10-4 and 10-5.
Artificial Data Distribution example

The "Mode" for this dataset should be set to "Permutation Values".

Artificial data generated from the normal distribution

Permutation values are obtained by randomly drawing samples from the normal distribution with zero-mean and unit-variance. This dataset contains 10000 permutation values for each of ten events/experiments. The correct theoretical permutation test P-values for these ten events are found when evaluating the cdf of the normal distribution at the values of the original statistic (on the first row in the dataset). These P-values are:10-1, 10-2, 10-3, 10-4, 10-5, 10-6, 10-7, 10-8, 10-9 and 10-10.

Artificial Data Normal example

The "Mode" for this dataset should be set to "Permutation Values".

Yeast gene expression data

The employed yeast data is the one mentioned in Section 3.2 of the paper. It contains 63 genes across 170 microarrays. In 80 of these arrays yeast was grown aerobically; for the other 90 arrays yeast was grown anaerobically; this division constitutes the labels. Permutation values are obtained by computing the SAM statistic on the yeast expression data using permuted label configurations; the original statistic is computed by using the original label configuration. The original dataset can be downloaded here.

This file also contains the correct permutation test P-values that are computed as explained in Section 3.2 of the paper. For each gene 10000 permutation values were generated.

• Yeast Data example
• Excel file format

The "Mode" for this dataset should be set to "Permutation Values".

SAM/GSEA for Response Type 'Two class unpaired'

• SAM/GSEA Data
• Smaller SAM/GSEA Data

The "Mode" for this dataset should be set to "SAM" or "GSEA".

GSEA gene set file (in .gmt format)

• geneset.gmt
• genesets_small.gmt