Hi, I have used DEXSeq for differential exon usage testing. I believe what DEXSeq uses as threshold for finding significant differentially expressed exons is p-adjust < 0.1 and it reports both p-value and p-adjust in its results. My question is that what is the difference between p-adjust and p-value? when I use p-adjjust < 0.1 as threshold, I get a few number of genes, while in the case of p-value < 0.05, I get ,much more genes. I was wondering which threshold is more accurate? Sincerely Yours, Delasa Aghamirzaie Genetics, Bioinformatics, and Computational Biology (GBCB) PhD Student Virginia Tech Blacksburg, Virginia [[alternative HTML version deleted]]

On Jun 19, 2013, at 1:10 PM, Delasa Aghamirzaie wrote: > Hi, > I have used DEXSeq for differential exon usage testing. I believe what > DEXSeq uses as threshold for finding significant differentially expressed > exons is p-adjust < 0.1 and it reports both p-value and p-adjust in its > results. > My question is that what is the difference between p-adjust and p-value? Dear Delasa, I am not sure which option p-adjust you used but if it is the Benjamini-Hochberg false discovery rate it estimates how many genes which were accepted as differentially expressed based on the cutoff are in fact not differentially expressed. Fdr is more stringent than p-value but does not have to have as low a value. Say you usea p-value cutoff <0.05. As long as this value is not exceeded it is resonable to take p-adjust (fdr) at 0.1 or even as high as 0.5 as long as you know that in the latter case have the genes you try to validate by pcr are expected not to be differentially expressed. I think that p-adjust<0.05-0.25 are good cutoffs. With hopes that the above helps, Rich Richard A. Friedman, PhD Associate Research Scientist, Biomedical Informatics Shared Resource Herbert Irving Comprehensive Cancer Center (HICCC) Lecturer, Department of Biomedical Informatics (DBMI) Educational Coordinator, Center for Computational Biology and Bioinformatics (C2B2)/ National Center for Multiscale Analysis of Genomic Networks (MAGNet)/ Columbia Initiative in Systems Biology Room 824 Irving Cancer Research Center Columbia University 1130 St. Nicholas Ave New York, NY 10032 (212)851-4765 (voice) friedman at cancercenter.columbia.edu http://cancercenter.columbia.edu/~friedman/ In Memroriam, John Holbrook Vance > when I use p-adjjust < 0.1 as threshold, I get a few number of genes, while > in the case of p-value < 0.05, I get ,much more genes. I was wondering > which threshold is more accurate? > > Sincerely Yours, > Delasa Aghamirzaie > Genetics, Bioinformatics, and Computational Biology (GBCB) PhD Student > Virginia Tech > Blacksburg, Virginia > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

Hi Delasa On 19/06/13 19:10, Delasa Aghamirzaie wrote: > I have used DEXSeq for differential exon usage testing. I believe what > DEXSeq uses as threshold for finding significant differentially expressed > exons is p-adjust < 0.1 and it reports both p-value and p-adjust in its > results. > My question is that what is the difference between p-adjust and p-value? > when I use p-adjjust < 0.1 as threshold, I get a few number of genes, while > in the case of p-value < 0.05, I get ,much more genes. I was wondering > which threshold is more accurate? The p.adjust column contains p values which have been adjusted for multiple testing with the Benjamini-Hochberg method. As you seem to be unfamiliar with these important concepts: A while ago, I tried to give an as-simple-as-possible explanation what this is about in this thread on SeqAnswers: http://seqanswers.com/forums/showthread.php?t=17011 Simon