Entering edit mode
8.4 years ago
biotech
▴
570
What's normally the procedure for the manual curation of bacterial annotations? We've thought about running different tools, such as RAST and Prokka and then compare annotations with tools such as BEACON.
However, I'm finding quite complicated to get Genbank formats ready to BEACON tool since they do not agree with what it accepts as the correct Genbank format.
Would you recommend a different procedure?
You need to load your genome and the corresponding annotation in a manual curation tool like webapollo. Then you load other tracks (information) allowing you to judge and modify (manually) the gene models. These other tracks could be transcripts alignment, protein alignment, RNAseq (bam files), CAGE data, etc.
That is not true manual curation. A full manual curation would require one to look at the predicted gene models, pore over every blast alignment, compare the domains/active site structure by building multiple sequence alignments before making a final call.
This is the Million dollar component of the $1000 genome.
If you are going to eventually make the data public put it through NCBI's prokaryotic annotation pipeline.