Gff3 annotation

[lupid]

Hi to everyone,

I'm working in peach, and in this model I have 2 very big problems in my RNA-Seq analysis.
1 Is imposible to find a GFT file
2 In the GFF3 file, the annotation is in a separate file, and
after I run the tophat/cufflinks suit, I have no annotation in the final file, please someone help me, how can I merge this GFF3 with de annotation file, provided by the same database.

[westerman]

Maybe you can give more details. I use GFF3 files with tophat/cufflinks without any problems. As for combining a GFF3 file with the output of cufflinks then using the gene identifiers should allow you to do this. It does depend on your GFF3 file. Most of this information is contained in the 9th field which is free-format and not well defined (although, as I recall, tophat does require certain sub-sections).

[GenoMax]

@lupid: Ensembl appears to have a better GFF3. Of course that may not match what you started with in terms of the reference genome. ftp://ftp.ensemblgenomes.org/pub/pla...5.1.27.gff3.gz

Are you using the files from (PeGDB: http://www.plantgdb.org/XGDB/phplib/...d.php?GDB=Pe)?

[lupid]

Thanks a lot, but what I'm looking for is the function of the gene, this is what we call annotations, as the gtf has

[westerman]

@lupid. Not all GTF files have annotations (and in fact GTF is the same as GFF2) so insisting on a GTF will not necessarily give you annotations. It all depends on who makes up the GTF/GFF file. If they put in annotation then great. If not then they still created a valid file.

As GenoMax hinted, you do need to use the GFF/GTF that has the same positions as your reference sequence. If you are using PeGDB then you could combine the information in the GFF3 and the annotation file to create an annotated GFF file. If you are using a reference from a different source then please tell us the source and we can give advice on how to create an annotated GFF file.