In the analysis of thousands of genes with only a small number of samples, a major challenge facing functional genomics is the assessment of statistical significance, that is, determining how many findings are false or the estimated rate of false-positive results. We describe a simple permutation method to evaluate exactly the genome-wide significance level, acknowledging both multiple comparisons and small sample sizes. We illustrate this method through an application to a study of leukemia.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Zhao, L., Thomas, J. Toward rigorous assessment of statistical significance in functional genomics. Nat Genet 27 (Suppl 4), 98 (2001). https://doi.org/10.1038/87382
Issue Date:
DOI: https://doi.org/10.1038/87382