[Version 10/1/05+] We can also use the "Analysis/Filter Genes/Percentile filtering" function to select genes by its fold change between a high and a low percentile across samples. The filtered gene list can be clustered and viewed as usual. This gene list file also contains the percentile-standardized data, with the two percentiles for each gene linearly scaled to the +/- displaying range (specified at “Options/Clustering”). To view the percentile-standardized data, first do 'Get external data' to read in this exported data file, and then do “Analysis/Clustering” with “Options/Standardize rows” unchecked.