How to Generate GO annotation from list of genes

Question

[madkitty]

We're currently working on a genome that doesn't have any GAF file for gene ontology with Ontologizer. In order to create the GAF file, for this genome, we need to generate GO annotation from either Interproscan or Blast2GO. We have a list of genes from NCBI (total of 2,000 genes) as below, I've been reading the manual of Blast2GO, but that looks like Chinese to me, How to use Blast2GO to generate GO annotation from a list of gene symbols??

5HTT
A3F2
ABCC1
ACCN1
ACCN3
ACP5
ACTB
ACTN3
ACVR2A
ADAMTS4
ADIPOR2
ADORA2A

Also, I'm very open minded to hear if you have an alternative to gene ontology for genomes that don't have any GAF file yet.

[st.ph.n]

Try inputting your gene list into DAVID. You can also use Maker, which you can help train with Interproscan, and you'll get a .gff3 file. You can run BLAST2GO two ways: java gui, or command line. The commandline is a little trickier if you're new to Linux or programs. On the GUI, run your BLAST separately, saving in XML format, as well as interproscan. Upload each results to the GUI, as well as your genome.

[bruce.moran]

Beat me to it!

[Adrian Pelin]

Is there a way to work with non-models when using DAVID? If I annotate something with blast2go, can I then use DAVID?

[st.ph.n]

Not in my experience. BLAST2GO gives you nice graphs to analyze, but not real GO terms. You may just want to go with Interproscan, to get GO terms and input that list into DAVID.

Answer 1

[bruce.moran]

You have gene names, so you can look at ontology of human orthologs, or other more relevant species but human has most annotations, obviously. Use DAVID, it gives most options for ontology, pathways, functional annotation clustering etc. Best to include a background of other genes that are expressed/have SNPs in or around them/whatever your study is looking at (that sort of info is good to include in questions=)

Answer 2

[EagleEye]

Try this: Gene Set Clustering based on Functional annotation (GeneSCF)