Intersection between Up and down regulated genes

[clbenoit]

trafficstars

I find an intersection between the up and down regulated genes with Ideal. ("Intersection of gene sets" figure in the "Functional analysis" outlet). That is strange. Please find the figure in PJ plot_vennlists.pdf

Best wishes, Cala3len

[federicomarini]

Indeed weird. Got a MRE I can work on?

[clbenoit]

Hello Frederico, Please find here both sampleplan and count matrix I have used : IdealData.zip (Filter 2 on the rowsums then group_1 vs group2 DE with default parameters).

[federicomarini]

Hi @Cala3len, I might need a more detailed description to fully reproduce this one. What should I specify in the other steps for the setup part? Or can you provide the exact code that underlies your sentences?

[federicomarini]

Although, from a first look at it, I have some first bets

• handling of NAs - might be not so completely correctly handled by gplots::venn
• might be handled differently by the upset plot function

Probably should be addressed by an extra na.omit ahead of the call.

Could you confirm that if you save the state of the reactives (upper right corner)? Or as well, if you are confident with the editor: try adding such a chunk while the app is live

values$genelistUP()
values$genelistDOWN()
intersect(values$genelistUP(),values$genelistDOWN())

sum(is.na(values$genelistUP()))
sum(is.na(values$genelistDOWN()))

vl <- gplots::venn(
  list(
    values$genelistUP(),
    values$genelistDOWN()
  )
)

vl