Rank products: A simple, yet powerful, new method to detect differentially regulated genes in replicated microarray experiments

FEBS Letters (2004) 573: 83-92.

Rainer Breitling (1,2), Patrick Armengaud (1), Anna Amtmann (1), and Pawel Herzyk (2,3)

(1) Plant Science Group,
(2) Bioinformatics Research Centre, and
(3) Sir Henry Wellcome Functional Genomics Facility
Institute of Biomedical and Life Sciences (IBLS)
University of Glasgow
Glasgow G12 8QQ
United Kingdom

{R.Breitling, P.Armengaud, A.Amtmann, P.Herzyk}@bio.gla.ac.uk

Erratum

In the example on the second page of the manuscript (Determination of significance levels) the calculation of the p'-value is not described correctly. Because the two RP values are not independent, they can't be added to give the combined probability. Instead of p' = 2RP = 4/9 the correct value is p' = 3/9. The remaining description in the example and the algorithm itself are not affected by this oversight.

In table 3, the numbers in the printed version are not correct for the RP method. A corrected version of the table is available for download. The apparent advantage of RP over SAM is even more striking with the revised numbers.

Back to Homepage.