Does anyone have an easy fix for the TAIR (Arabidopsis thaliana) "GFF3" files. I've used all the GFF3 validators I can find online and they are all telling me that they are riddled with formatting errors. I've tried the .gff files from a number of different genome versions from TAIR, none of them work e.g. TAIR10_GFF3_genes.gff
-
-
What are you trying to use them with? I have not had any problems. TAIR does use some unconventional naming, but the formats work fine. -
Trying to use with a Perl based pipeline, "GENE-counter". I'll keep looking for a fix, but it's a little dissapointing that the TAIR sequence references don't event match between the .gff and genome.fa...
Thanks for the replyComment
-
The only sequences that don't match that I am aware of is the Mitochondria and Chloroplast names. In the gff these are named as "Mitochondria" and "Chloroplast" while in the genome.fa they are listed as "ChrM" and "ChrC" respectively (or the other way, its been a while since I looked at it). Chr1-5 should be identical.
Given that Arabidopsis only has the 5 chromosomes plus mitochondria and chloroplast, the easiest solution is to just go into the genome.fa and change the names of the chromosomes to match the gff. It takes all of 10 seconds and usually solves most discrepencies.Last edited by chadn737; 04-20-2013, 11:31 AM.Comment
-
Yes, I've had to change those as you describe. I've managed to work around some of the GENE counter issues, which I think might have been the bigger problem. Thanks again for the input!Comment
-
There are a few gene names containing semicolons in the TAIR GFF files which confuse a lot of parsers. Removing these might help.Comment
-
I'll look into that, thanks.Comment
-
We discuss how to fix at least one set of problems with the TAIR version of GFF files here:
If you don't need to convert TAIR GFF to BED, you could just pipe the result of the GNU awk statement to a GFF file, e.g.:
$ awk '{gsub(/;$/,"");print}' TAIR9_GFF3_genes.gff > TAIR9_GFF3_genes.fixed.gffComment
-
Researchers using sequencing and genomics tools often have to make trade-offs. They can choose between speed or scale, short reads or long-range information, or targeted panels or a view of the whole transcriptome. New technologies that have been released this year are built to address those tough choices.
We asked six companies the same four questions to learn about their latest products. The new technologies bring a lot to the table, including rethinking sequencing...
Comment