Hello Sir, I am using it for non-model organisms.

1) I have a set of differentially expressed genes, plus the go term related to genes. 2 ) The command in R in need to use is: goseq(pwf, genome, id, gene2cat = Null) I understand that for gene2cat argument I need to mention my list of genes with their go term.Correct me if I am wrong. For pwf= I need to use nullp(DEgenes, genome, id, bias.data=NULL), however what should I use for genome and id here, and for bias.data (or can i skip genome and id part). I know length of each gene, can that be used for bias.data

3)Lastly for genome, id again, what should I mention.

I am guessing I can skip the genome and id part here and just provide two: argument pwf and gene2cat.

Thanking you in advance for your help.

There are two ways you can get mappings between your gene IDs and GO terms. One, you can tell goseq what genome and ids you used, in which case it will do the mappings for you, or two, you can provide a gene2cat data.frame or list that will be used instead.

If your IDs are NCBI Gene IDs, it's just as easy to specify the genome and id arguments. But if you are using a non-model organism, or if your IDs are Ensembl IDs, I would personally recommend generating your own gene2cat object and using that instead of the automatic mapping. For example, if you have Ensembl IDs, I would use the biomaRt package to get the Ensembl ID to GO mappings, rather than having goseq do the mapping for you.

Also, do note that Null is not the same as NULL.