Dear List I've some doubts about the usage of a function included in oneChannelGUI package: According to the oneChannel manual I look that after generating the linear model it's possible to evaluate the possibility to correct the computed p-values with the function raw p-value distribution. If the p-values distribution is not uniform in the non significant range is not possible to apply any correction. This is not completely clear to me and I'm not able to find more documentation about the general topic of when to correct or not the p-values. For example examining the raw p-value plots of some microarray experiments I've some cases on which the raw distribution of my p-values is quite uniform in the not significant range but the values in this range are more abundant that the values in the significant range. In this case the best approach is correcting my p values (even if probably I won't obtain significant results) or not correcting them? Thank you very much for every tips you could give to me Guido [[alternative HTML version deleted]]