Hi everyone, Apologize first for probably simple question below. Thanks for Jim's help, I've got some html files from limma2annaffy function, however, I am not pretty sure how to interpret the result of "fold change". Please correct me if I am wrong. Based on my knowledge, after limma, for example in coef=1, the fold change represents the contrast of the expression value means from two groups. But the html result made me confused: The header of html is like this: probes Description Pubmed Gene Ontology Pathway t-statistic P-value Fold-change array1 array1 array2 array2 The question is why the fold change is not really the value that (mean of group1)- (mean of group2) ?? If it is because of permutation?? I've read limma user guide and ?limma2annaffycore, so I believe the value of array 1 and 2 are expression value. By the way, the function of pflt filter genes by p value, but if I want to filter gene by adj.p.val?? Thank you very much!! Hui-Yi [[alternative HTML version deleted]]

Hi Hui-Yi, Hui-Yi Chu wrote: > Hi everyone, > > Apologize first for probably simple question below. > Thanks for Jim's help, I've got some html files from limma2annaffy function, > however, I am not pretty sure how to interpret the result of "fold change". > Please correct me if I am wrong. Based on my knowledge, after limma, for > example in coef=1, the fold change represents the contrast of the expression > value means from two groups. But the html result made me confused: > > The header of html is like this: > probes Description Pubmed Gene Ontology Pathway t-statistic > P-value Fold-change array1 array1 array2 array2 > > > The question is why the fold change is not really the value that (mean of > group1)- (mean of group2) ?? If it is because of permutation?? The fold change *is* mean(goup1) - mean(group2). If you show us your code and a snippet of the output we might be able to help (please see the posting guide). > I've read limma user guide and ?limma2annaffycore, so I believe the value of > array 1 and 2 are expression value. > By the way, the function of pflt filter genes by p value, but if I want to > filter gene by adj.p.val?? I don't know of any function pflt. However, if you mean the argument pfilt, then it *does* filter by adjusted p-value if in fact you are adjusting the p-value to begin with. Best, Jim > > > Thank you very much!! > Hui-Yi > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- James W. MacDonald, M.S. Biostatistician Hildebrandt Lab 8220D MSRB III 1150 W. Medical Center Drive Ann Arbor MI 48109-0646 734-936-8662