ChipST2C is a Windows XP computer software package which identifies statistically differentially expressed DNA-microarray based genes among k independent samples. The statistical tests employed are the parametric t-test and non-parametric Mann-Whitney U test when there are k=2 samples, and the parametric F-test or non-parametric Kruskal-Wallis test for k>2 samples. All four tests (i.e., t, M-W, F, and K-W) can be carried out as a permutation-based randomization test or without randomization. In addition, either a "within-gene" permutation or "between-gene" permutation scheme can be specified. Also, for the multiple testing problem, ChipST2C generates Bonferroni adjustments to p-values, the Benjamini & Hochberg false discovery rate (FDR) method, or positive false discovery rate (pFDR) method reflected in Storey q-values.
Significant genes are then displayed automatically in cluster images using agglomeration and distance function specified by the user. In addition, significant genes and their test results are sorted by either p-value or test statistic and written to disk in the form of Excel .csv files. The user can also perform a k-means cluster analysis of all of the genes in a single cluster or genes with q-values (or p-values) below a given level. The k number of clusters can also be determined automatically or fixed by the user prior to run-time.
