Blast2go input files

[Bioinformaticsnewb]

Hi everyone,

I have a question regarding the options for blast2go input. I am currently analyzing an RNAseq project comparing gene expression patterns during four treatments, each with three biological replicates. I am using the Tuxedo suite (TopHat, Cufflinks, Cuffmerge, Cuffdiff) for assembly and differential expression analysis and I would like to look at the ontology of the differentially expressed genes that I was able to pull out from the cuffdiff output, but it doesn't seem to be in a format that blast2go can deal with. Has anyone else run into a similar issue? And if so, how do I go about getting my GO data for these genes? Is a file format change necessary, or am I missing something?

Thanks,
Alex

[maubp]

Blast2GO just needs a FASTA file of your gene sequences (one record per gene), or a BLAST XML file of your genes searched against the NCBI NR database.

[Bioinformaticsnewb]

Blast2GO just needs a FASTA file of your gene sequences (one record per gene), or a BLAST XML file of your genes searched against the NCBI NR database.

Thanks for the input. I am aware that a FASTA file is required for blast2go, but I am primarily interested in annotating only the transcripts that were differentially expressed among the treatment groups. Since that information in contained in the cuffdiff output, I'm curious if there is a way to generate a fasta file that contains only those sequences of interest.

Thanks again,
Alex

[maubp]

In that case just filter your FASTA file of all genes to make a new smaller FASTA file of just the genes of interest (using the differentially expressed gene ID list from cuffdiff). That should be trivial in your scripting language of choice.

[Jeremy]

Blast2GO has a GO term enrichment function, that might be what you want. You need a Blast2GO result for all sequences then give the test set as a list of sequence names that is then compared against all the sequences.