Thanks for the reply Steve!

So I spent literally all day trying to figure out how to go about doing that, now that I know its okay...and I think maybe I finally figured it out? Getting the data into the right format for GOseq is a bit of a beast for a novice R user!!

Any chance you're feeling super friendly and could take a look at my code?

Also, just in case someone in the future needs to do something like this, maybe the code will help them avoid a day of banging their head on the wall (assuming its correct code :D).

When I run GOseq the pwf plots are looking a bit odd which is the only thing that has me worried... This is link to the pwf plot https://ibb.co/h69iLH

#Get transcript lengths from Tximport data
txi.length.df <- as.data.frame(txi[["length"]])

#Subset for only genes input to the DeSeq DGE model
txi.length.df <- subset(txi.length.df, rownames(txi.length.df) %in% rownames(results.df))

#Remove samples not being included in analyses at this time
txi.length.df_sub <- within(txi.length.df, rm("SRR000020643", "SRR000020650", "SRR000020653", "SRR000020660", "SRR000020663", "SRR000020670", "SRR000020680", "SRR000020690", "SRR000026811"))

#Calculate mean tx length
txi.length.df_sub$avgTx <- rowMeans(txi.length.df_sub)

#Load file containing the list of genes DE from DeSeq2 output
DE.genes <- read.delim(file = "/users/rimlandca/desktop/RNA Seq Results/GOseq files/DEGs/Upreg overlap.txt")

#Load file containing GO terms (downloaded GO terms from Ensembl Biomart online for protein coding genes)

go.terms <- read.csv("/users/rimlandca/desktop/RNA Seq Results/GOseq files/Category Mappings/GO term.txt", as.is=TRUE)
head(go.terms)
names(go.terms) <- c("Gene.stable.ID","GO.term.accession")
summary(go.terms)

#Background Genes
expressed.genes <- row.names(txi.length.df_sub)

#Gene Length
gene.lengths <- txi.length.df_sub$avgTx

#Make vector where DEGs = TRUE and non-DEGs = FALSE

DEgenes <- expressed.genes %in% DE.genes$ID
names(DEgenes) <- expressed.genes
head(DEgenes)
sum(DEgenes)

#Run GoSeq!

nullp.result <- nullp(DEgenes = DEgenes, bias.data = gene.lengths)

GO <- goseq(pwf = nullp.result, gene2cat = go.terms)

write.table(GO[GO$over_represented_pvalue < 0.01],row.names=FALSE,file="GO_terms.txt", quote = FALSE,col.names = FALSE)