Email updates

Keep up to date with the latest news and content from BMC Bioinformatics and BioMed Central.

Open Access Methodology article

False discovery rate control in two-stage designs

Sonja Zehetmayer1* and Martin Posch2

Author Affiliations

1 Center for Medical Statistics, Informatics, and Complex Systems, Medical University of Vienna, Vienna, Austria

2 European Medicines Agency (EMA), London, UK

For all author emails, please log on.

BMC Bioinformatics 2012, 13:81  doi:10.1186/1471-2105-13-81

Published: 6 May 2012

Additional files

Additional file 1:

We report the simulation scenarios and results of the simulation study assessing the FDR of the FNS and FDRS design (modified procedure with ms=6) for the case of independent test statistics as described in the results section of the manuscript. For each scenario at least 1000 simulation runs were performed. For scenarios with lower m the simulation runs were increased to 50000 (m = {100;500}), 20000 (m = 1000), and 10000 (m = 5000), because in these scenarios there is a higher variability of the false discovery proportion such that the estimator of the FDR converges slower. This also holds if m is large but π0 ≈ 1 or Δ is small. Therefore, for these scenarios the number of simulation runs was increased. The resulting FDR values were plotted as a function of α1 for the FDRS design (left column) or as a function of m2 for the FNS design (right column), respectively.

Format: PDF Size: 106KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 2:

Results of a simulation study for two-stage designs where an adaptive test procedures is applied based on an estimator for the proportion of true null hypotheses.

Format: PDF Size: 106KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 3:

Two single-stage designs are compared to the results: For the first single-stage design the sample size for each hypothesis is n1, for the second design the sample size is n1 + n2. For the first design we compare the gain in power of the integrated design and for the second design the attention lies on the reduction in costs.

Format: PDF Size: 106KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data