There's an option in MRfulltable / MRtable called ncounts. Please see the help file for more info. On 2018-01-31 01:46, arguello.rguez [bioc] wrote: > Activity on a post you are following on support.bioconductor.org [1] > > User arguello.rguez [2] wrote Question: Filter low counts OTUs from a MRfull table to prepare a heatmap [3]: > > Hi, > > I am building heatmaps from the MRfulltable outputs obtained in metagenomeseq. When preparing the heatmap I am able to filter OTUs by adjusted p-value, so I can prepare a heatmap chossing only the ssignificant OTUs, however it would be also desirable to remove those significant OTUs with low count numbers.... When I try to use a similar code, lets say > > "significant_OTU <- which(MRfulltable(Donor_A, number=length(Donor_A$taxa))"$countsingroup1>100" > > I get an error, > > Any advice? > > I paste my code below, thanks in advance for the response and also for your time. > > heatmapColColors = c("grey","green")[as.integer(factor(pd_Donor$Donor))] > > heatmapCols = colorRampPalette(c("white","light blue","blue","dark blue"))(50) > > significant_OTU <- which(MRfulltable(Donor_A, number=length(Donor_A$taxa))$ADJPVALUES<=0.05) > plotMRheatmap(obj = metHMEX6_A[significant_OTU,] , n=length(significant_OTU), margins=c(4,14),cexRow = 0.7, cexCol = 1, trace = "none", col = heatmapCols, ColSideColors = heatmapColColors, lhei = c(2, 8)) > > ------------------------- > > Post tags: metagenomeseq > > You may reply via email or visit Filter low counts OTUs from a MRfull table to prepare a heatmap Links: ------ [1] https://support.bioconductor.org [2] arguello.rguez [3] Filter low counts OTUs from a MRfull table to prepare a heatmap