Statistics / Sample size calculations with an adapted BH method

Description

Perform sample size calculations using an adapted Benjamini-Hochberg method.

Parameters

• Phenodata column separating the samples into two groups
• Assume equal variances between groups (yes, no) [no]
• Distribution to calculate p-values from (normal, student) [normal]
• False discovery rate (0-1) [0.1]
• Image width (200...3200) [600]
• Image height (200...3200) [600]

Details

This tool uses an adapted version of the Benjamini-Hochberg method to perform sample size calculations using a pilot data set. The necessary parameters, such as effect size and the proportion of differentially expressed genes, are estimated from the data. For more details, please see the referenced articles.

Output

A plot and a text table showing estimates of average power as a function of sample size.

A set of quality control plots. For interpretation, please see the referenced CGHpower paper, and these evaluation data sets.

References

Technical description of the method:
Ferreira et al. (2006) Approximate power and sample size calculations with the Benjamini-Hochberg method. Int J Biostat 2: Article8

An illustration with real and simulated microarray expression data:
Ferreira et al. (2006) Approximate sample size calculations with microarray data: an illustration.. Stat Appl Genet Mol Biol 5: Article25

The use of the method in the context of aCGH (CGHpower). Whereas the other two references are fairly technical, this one is written mostly for biologists:
Scheinin et al. (2010) CGHpower: exploring sample size calculations for chromosomal copy number experiments. BMC Bioinformatics 11: 331