Hi, I implemented filtering by co-efficient of variation (CV) approach to filter my array data, in order to reduce the problem for multiple testing. In order to come up with a threshold to filter by CV, and to support the final result ( i.e saying "certain value of CV is proved statistically is ideal to get true differentially expressed genes) . I am trying to calculate true positives(TP) and false positives(FP) in my differentially expressed gene list, obtained after filtering with different cut-off's, and plot ROC curves to see the power of filtering by CV, against FDR. Is there any inbuilt function in R to calculate the TP/FP ? If not, can any one give me an initial start , ( i am not a statistician, but interested to learn the concepts, so need an initial start ) I am stuck with the point in choosing a standard/ control in order to classify my output to TP and FP. Any other suggestions will be appreciated. thankyou sundar [[alternative HTML version deleted]]