topGo analysis getting same number of Annotated, significant and expected

Question

[rashmi.jain20]

Hi,

I am trying to rum topGo analysis for some of the DE genes using topGO, but I am getting the same number of annotated, significant and expected.

Below is my code:

library(topGO)
library(org.At.tair.db)

tmp<-read.csv("./topGo-edgeRAll-input.csv")
geneList<-tmp$FDR
names(geneList) <-tmp$Gene

head(geneList)
GOdata <- new("topGOdata",
              ontology="BP",
              allGenes=geneList,
              geneSelectionFun=function(x)(x==1),
              annot=annFUN.org, mapping="org.At.tair.db")

resultKS<- runTest(GOdata, algorithm = "weight01", statistic = "ks")
resultKS
tab<- GenTable(GOdata, raw.p.value=resultKS, topNodes=length(resultKS@score), numChar=120)
head(tab)

Getting results:

GO.ID                                                 Term Annotated Significant Expected raw.p.value
1 GO:0009718 anthocyanin-containing compound biosynthetic process        14          14       14     1.9e-06
2 GO:0009744                                  response to sucrose        23          23       23     3.1e-05
3 GO:0010224                                     response to UV-B        20          20       20      0.0002
4 GO:0080167                                 response to karrikin        17          17       17      0.0016
5 GO:0002679       respiratory burst involved in defense response         6           6        6      0.0022
6 GO:0006857                               oligopeptide transport        10          10       10      0.0081

I appreciate your help!

Best

Answer 1

[Kevin Blighe]

I think that it's due to the fact that you are using the 'weight01' algorithm, just as they use here: https://ucdavis-bioinformatics-training.github.io/2018-June-RNA-Seq-Workshop/friday/enrichment.html

Also, be very careful about the use of this parameter: geneSelectionFun - in your code, you are only retaining genes with FDR q-values (adjusted p-values) == 1, as this example shows:

genes
    a     b     c     d     e 
1.000 0.500 0.001 1.000 6.000 

geneSelectionFun=function(x)(x==1)
genes[geneSelectionFun(genes)]
a d 
1 1

If you want to automatically include all genes, then use:

geneSelectionFun = function(x) (TRUE)

Kevin