gffutils error for bed12 conversion in scikit-ribo-build.py

[drewjbeh]

Hi there,

I am trying to test scikit-ribo on some yeast ribo-seq data we have just generated.
When running scikit-ribo-build.py I get the following error:
File "/home/drew/anaconda3/envs/scikit-ribo/lib/python3.6/site-packages/gffutils/interface.py", line 227, in __getitem__ raise FeatureNotFoundError(key) gffutils.exceptions.FeatureNotFoundError: YDR106W_BY4741

I am using the yeast BY4741 Toronto genome, and a subset of the associated gff file that I converted to gtf using gffread. The subset just includes coding genes (and excludes some other Dubious annotated transcripts). This seemed to solve another issue I was having using the full gff or gtf made from this which were either
pandas.errors.ParserError: Too many columns specified: expected 9 and found 1.
when using a gff, or
KeyError: 'gene_id'
when using a gtf that included noncoding genes (the problem here is with missing "gene_id" descriptors for these kinds of genes or for exon annotations.)
However I ran into the problem described above.

I am not sure why gffutils is not finding the genes by these names (e.g. YDR106W_BY4741) since this is exactly how they are named in the gtf file (obviously, since the geneNames list in gtf_preprocess.py is built from the gtf itself).

Your help would be much appreciated!! Thanks.

[mschatz]

Thanks for your interest! Does the yeast gtf file that is included in the examples directory work? Even if you want to switch to a different reference, checking that one first would help size up exactly where the error is introduced. And Im a little bit worried that it could be as you are removing/changing transcripts from the gene annotation that the rnafold results will also have to be adjusted as well. Cheers, Mike

…

On Thu, Apr 5, 2018 at 5:31 AM, drewjbeh ***@***.***> wrote: Hi there, I am trying to test scikit-ribo on some yeast ribo-seq data we have just generated. When running scikit-ribo-build.py I get the following error: File "/home/drew/anaconda3/envs/scikit-ribo/lib/python3.6/ site-packages/gffutils/interface.py", line 227, in __getitem__ raise FeatureNotFoundError(key) gffutils.exceptions.FeatureNotFoundError: YDR106W_BY4741 I am using the yeast BY4741 Toronto genome, and a subset of the associated gff file that I converted to gtf using gffread. The subset just includes coding genes (and excludes some other Dubious annotated transcripts). This seemed to solve another issue I was having using the full gff or gtf made from this (pandas.errors.ParserError: Too many columns specified: expected 9 and found 1.) However I ran into the problem described above. I am not sure why gffutils is not finding the genes by these names (e.g. YDR106W_BY4741) since this is exactly how they are named in the gtf file (obviously, since the geneNames list in gtf_preprocess.py is built from the gtf itself). Your help would be much appreciated!! Thanks. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#4>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAL982meO5_4_mIwA9cw5Pdnb5uR45d0ks5tleR1gaJpZM4TIIly> .