Dear R-Help, I am using the function "topTable" from the LIMMA package. To estimate adjusted P-values there are several options (adjust="fdr" , adjust="BH") as shown below: topTable(fit, number = 10, adjust = "BH", fit$Name) I guess any of these options (fdr, BH, etc.) is using a default of FDR=0.05 which is quite conservative (i.e., very likely none of the tested genes will be ranked as differentially expressed at 5% FDR). I would argue that for an exploratory study it would be okay to use a 0.05 < FDR <0.20. So, I was wondering if we can change that default of FDR=0.0.5 with a less stringent FDR value (0.10, 0.20, etc.) in the "topTable" function. How do I change it? Thanks a lot for the help on the use of "topTable" function from the LIMMA package. Roger Roger L. Vallejo, Ph.D. Computational Biologist & Geneticist U.S. Department of Agriculture, ARS National Center for Cool & Cold Water Aquaculture 11861 Leetown Road Kearneysville, WV 25430 Voice: (304) 724-8340 Ext. 2141 Email: roger.vallejo@ars.usda.gov <mailto:roger.vallejo@ars.usda.gov> [[alternative HTML version deleted]]
Hi Roger, Vallejo, Roger wrote:> Dear R-Help, > > > > I am using the function "topTable" from the LIMMA package. To estimate > adjusted P-values there are several options (adjust="fdr" , adjust="BH") > as shown below: > > > > topTable(fit, number = 10, adjust = "BH", fit$Name) > > > > I guess any of these options (fdr, BH, etc.) is using a default of > FDR=0.05 which is quite conservative (i.e., very likely none of the > tested genes will be ranked as differentially expressed at 5% FDR). I > would argue that for an exploratory study it would be okay to use a 0.05 > < FDR <0.20. So, I was wondering if we can change that default of > FDR=0.0.5 with a less stringent FDR value (0.10, 0.20, etc.) in the > "topTable" function. How do I change it?The topTable() function doesn't have a default cutoff based on the p-value (adjusted or otherwise). The cutoff is based on the 'number' argument, which has a default of 10, so regardless of the p-value adjustment method used, you only get 10 genes if you don't change this default. You can either increase the number to something arbitrarily high and then subset the resulting data.frame based on the p-value, or just use write.fit() to ouput the whole table and do the work in e.g., Excel. Best, Jim> > > > Thanks a lot for the help on the use of "topTable" function from the > LIMMA package. > > > > Roger > > > > > > Roger L. Vallejo, Ph.D. > > Computational Biologist & Geneticist > > U.S. Department of Agriculture, ARS > > National Center for Cool & Cold Water Aquaculture > > 11861 Leetown Road > > Kearneysville, WV 25430 > > Voice: (304) 724-8340 Ext. 2141 > > Email: roger.vallejo at ars.usda.gov <mailto:roger.vallejo at ars.usda.gov> > > > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.-- James W. MacDonald, M.S. Biostatistician Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 ********************************************************** Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.
Dear Roger, I think there is no a default value for the FDR in the toptable funtion. You will get a p-values corrected by the method you choose(fdr, bh). Briefly the shorted list of p-values will give you the FDR. It is up to you choose what is the estimated number of false positives you want to have. HTH Manuel --- "Vallejo, Roger" <Roger.Vallejo at ARS.USDA.GOV> escribi?:> Dear R-Help, > > > > I am using the function "topTable" from the LIMMA > package. To estimate > adjusted P-values there are several options > (adjust="fdr" , adjust="BH") > as shown below: > > > > topTable(fit, number = 10, adjust = "BH", fit$Name) > > > > > I guess any of these options (fdr, BH, etc.) is > using a default of > FDR=0.05 which is quite conservative (i.e., very > likely none of the > tested genes will be ranked as differentially > expressed at 5% FDR). I > would argue that for an exploratory study it would > be okay to use a 0.05 > < FDR <0.20. So, I was wondering if we can change > that default of > FDR=0.0.5 with a less stringent FDR value (0.10, > 0.20, etc.) in the > "topTable" function. How do I change it? > > > > Thanks a lot for the help on the use of "topTable" > function from the > LIMMA package. > > > > Roger > > > > > > Roger L. Vallejo, Ph.D. > > Computational Biologist & Geneticist > > U.S. Department of Agriculture, ARS > > National Center for Cool & Cold Water Aquaculture > > 11861 Leetown Road > > Kearneysville, WV 25430 > > Voice: (304) 724-8340 Ext. 2141 > > Email: roger.vallejo at ars.usda.gov > <mailto:roger.vallejo at ars.usda.gov> > > > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, > reproducible code. >