Ways to merge GO terms from different analyses?

Question

[nancydong20]

Hello!

I would like to do gene ontology annotation of a set of proteins from a non-model organism. I have been using BLAST2GO, mainly because it allows me to merge GO terms from the "mapping" step with those from InterProScan quite easily. However, it conducts web-based InterProScan and so is very slow. It seems that the free version of BLAST2GO does not allow me to import .xml result file from InterProScan v5.x (only .xml output from InterProScan v4.8), so I cannot run InterProScan separately. Because my organism of interest is pretty distant from mammals, InterProScan results are quite useful.

Is there a way to conduct GO mapping both through BLAST-based and InterProScan-based approaches, and then merge the GO terms for each gene?

Thank you very much!

[BioinfGuru]

What exactly do you mean by "merge"? Please give us an example input and an example output for a single gene.

Answer 1

[Whoknows]

I prefer to use Trinotate because it is much faster and free to use. You could merge GO data and Interproscan with Trinotate tool, it consists multiple database for analysis includes Interproscan, in final step it merge all databases result for your transcripts and gives those GO you want.

My own pipeline for GO analysis without reference genome:

1- Blastx (- outfmt 6 -max_target 1) transcripts with Swiss-Prot protein (from UniProt download section)

2-Use Target uniprot IDs and then send it to dcGO (http://supfam.org/SUPERFAMILY/cgi-bin/dcpredictormain.cgi) , this site generates GO, Pathway and .... based on protein sequence or UniProt IDS.

[nancydong20]

I'm sorry for the late reply, and thank you very much for the suggestion!

After looking at Trinotate, I see that it handles transcriptome data? What I have is the proteome set from a new genome, and would like to analyze at the protein level. Is there a program that can be recommended for that?