Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2006 Mar 2:7:106.
doi: 10.1186/1471-2105-7-106.

A simple method for assessing sample sizes in microarray experiments

Robert Tibshirani  1
Affiliations
Comparative Study

A simple method for assessing sample sizes in microarray experiments

Robert Tibshirani. BMC Bioinformatics. .

Abstract

Background: In this short article, we discuss a simple method for assessing sample size requirements in microarray experiments.

Results: Our method starts with the output from a permutation-based analysis for a set of pilot data, e.g. from the SAM package. Then for a given hypothesized mean difference and various samples sizes, we estimate the false discovery rate and false negative rate of a list of genes; these are also interpretable as per gene power and type I error. We also discuss application of our method to other kinds of response variables, for example survival outcomes.

Conclusion: Our method seems to be useful for sample size assessment in microarray experiments.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Results for simulated data. The genes are generated independently. Each panel shows the estimated FDR and FNR (solid red and green curves) as well as the 10 and 90th percentiles, using the proposed method (remember that in our setup FDR = 1-power and FNR = type I error). A horizontal line is drawn at 0.05. The quantity on the horizontal axis – number of genes – refers to both the hypothesized number of truly non-null genes, and the number of genes called significant. We see that the FDR is probably too high for the pilot data sample size of 20, but improves considerably when the sample size is doubled to 40.
Figure 2
Figure 2
Results for first simulation study. Here the FDR and FNR are estimated by direct simulation from underlying model.
Figure 3
Figure 3
Results for second simulated example (correlated genes).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Gilbert Chu, Balasubramanian Narasimhan, Robert Tibshirani, Virginia Tusher. Significance analysis of microarrays (sam) software http://www-stat.stanford.edu/~tibs/SAM/ via the Internet. Accessed 2003 July 16 2002.
    1. Lee M-LT, Whitmore GA. Power and sample size for microarray studies. Statistics in Medicine. 2002:3543–3570. doi: 10.1002/sim.1335. - DOI - PubMed
    1. Li SS, Bigler J, Lampe JW, Potter JD, Feng Z. Fdr-controlling testing procedures and sample size determination for microarrays. Statistics in Medicine. 2005:2267–2280. doi: 10.1002/sim.2119. - DOI - PubMed
    1. Muller P, Parmigiani G, Robert C, Rousseau J. Optimal sample size for multiple testing: the case of gene expression microarrays. J Amer Statist Assoc. 2005;99:990–1001. doi: 10.1198/016214504000001646. - DOI
    1. Pawitan Y, Michiels S, Koscielny A, Gusnanto S, Ploner A. False discovery rate, sensitivity and sample size for microarray studies. Bioinformatics. 2005:3017–24. doi: 10.1093/bioinformatics/bti448. - DOI - PubMed

Publication types

MeSH terms

LinkOut - more resources