Run topGO on non-model organism

topGO_nonModel_organism.R

library(dplyr)
library(ggplot2)
library(reshape2)
library(topGO)
library(plyr)

# Read in some table with gene ids, for example expressions
coreGenes <- read.table(<some_table_gene_ids>)

# Get gene mappings to GO terms
# This table has to look like this:
# FN03409	GO:0009514,GO:0004451,GO:0046872,GO:0006097,GO:0006099,
# FN23295	GO:0009507,GO:0022626,GO:0009514,GO:0005739,
# FN06463	GO:0009514,GO:0004474,GO:0006097,GO:0006099,
# FN16456	GO:0048046,GO:0009507,GO:0009570,GO:0005829,GO:0016020,GO:0005777,
# FN00210	GO:0005576,GO:0003993,GO:0046872,
# FN00543	GO:0005788,GO:0003756,GO:0045454,
# FN26703	GO:0005829,GO:0016020,GO:0005730,GO:0005634,GO:0005524,GO:0004612,
# FN16210	GO:0005829,GO:0016020,GO:0005739,GO:0005634,GO:0005886,GO:0009506,
# FN20048	GO:0009507,GO:0016020,GO:0005777,GO:0009536,GO:0004069,GO:0080130,
# FN26629	GO:0022625,GO:0003723,GO:0003735,GO:0000027,GO:0006412,

geneID2GO <- readMappings("gene2GO.txt")
geneNames <- names(geneID2GO)

# Get the list of genes of interest
myInterestingGenes <- coreGenes$id
geneList <- factor(as.integer(geneNames %in% myInterestingGenes))
names(geneList) <- geneNames
head(geneList)

GOdata <- new("topGOdata", ontology = "BP", allGenes = geneList,
                    annot = annFUN.gene2GO, gene2GO = geneID2GO)
# Run topGO with elimination test
resultTopGO.elim <- runTest(GOdata, algorithm = "elim", statistic = "Fisher" )
allRes <- GenTable(GOdata, elimKS = resultTopGO.elim,
                   orderBy = "elimKS", 
                   topNodes = 356)
write.table(allRes, file = "topGO_results.txt", sep = "\t", quote = F, col.names = T, row.names = F)