tpWY stands for "t-statistics, p-values, and Westfall and Young step-down method". It consists of software, developed at the Computational Biology and Informatics Laboratory (in the Center for Bioinformatics at the University of Pennsylvania) to implement the methods illustrated by Dudoit et al. (2000) to compute unadjusted and adjusted p-values for the t-statistics derived from a collection of two-channel microarray experiments (divided into two groups: the "treatment" and the "control" group). This paper is downloadable here. Albeit the application given in this paper is for data from two-channel microarrays, the same method can be utilized for other kinds of array data (filter arrays, oligonucleotide arrays). The method employs a step-down correction for multiple testing which follows Algorithm 4.1 in Westfall and Young (1993).
Besides an implementation of the algorithm described on page 14 of Dudoit et al. (2000) (where the alternative hypotheses are two-sided), the tpWY package also contains an implementation of the analogous algorithm for one-sided alternative hypotheses. The input file to these programs can consist of (normalized and possibly log transformed) data from mRNA expression experiments on any array platform.
To download the C source code, the makefiles, and the
usage documentation (README file) for the tpWY software click
If you have questions on the programs or their usage or if you want to report any bugs, please contact Elisabetta Manduchi.
Dudoit S., Yang Y.H., Callow M.J., Speed T.P. (2000) Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments. UC Berkeley, Technical report #578 .
Westfall P.H. and Young S.S. (1993) Resampling-based
multiple testing: examples and methods for p-value adjustment. Wiley
series in probability and mathematical statistics, Wiley.