Hi,

I received the following errors when running dexseq_count.py. I've flattened Mus_musculus.GRCm38.83.gtf downloaded from Ensembl (same errors on flattened Mus_musculus.GRCm38.75.gtf, Mus_musculus.NCBIM37.66.gtf). Should I report these errors to HTSeq instead? Thank you!

$ python dexseq_count.py -p no -s no accepted_hits.sam flattened.gtf dexseq.out
Traceback (most recent call last):
  File "dexseq_count.py", line 94, in <module>
    for f in  HTSeq.GFF_Reader( gff_file ):
  File "/usr/local/lib/python2.7/dist-packages/HTSeq/__init__.py", line 208, in __iter__
    ( attr, name ) = parse_GFF_attribute_string( attributeStr, True )
  File "/usr/local/lib/python2.7/dist-packages/HTSeq/__init__.py", line 164, in parse_GFF_attribute_string
    raise ValueError, "Failure parsing GFF attribute line"
ValueError: Failure parsing GFF attribute line

$ pip show HTSeq
---
Name: HTSeq
Version: 0.6.1p1
Location: /usr/local/lib/python2.7/dist-packages
Requires:

$ cat /path/to/DEXSeq/DESCRIPTION
Package: DEXSeq
Version: 1.16.7

----------------------------

EDIT:

$ head dexseq.gtf
1       dexseq_prepare_annotation.py    aggregate_gene  3073253 3074322 .       +       .       gene_id "ENSMUSG00000102693"
1       dexseq_prepare_annotation.py    exonic_part     3073253 3074322 .       +       .       transcripts "ENSMUST00000193812"; exonic_part_number "001"; gene_id "ENSMUSG00000102693"
1       dexseq_prepare_annotation.py    aggregate_gene  3102016 3102125 .       +       .       gene_id "ENSMUSG00000064842"
1       dexseq_prepare_annotation.py    exonic_part     3102016 3102125 .       +       .       transcripts "ENSMUST00000082908"; exonic_part_number "001"; gene_id "ENSMUSG00000064842"
1       dexseq_prepare_annotation.py    aggregate_gene  3205901 3671498 .       -       .       gene_id "ENSMUSG00000051951"
1       dexseq_prepare_annotation.py    exonic_part     3205901 3206522 .       -       .       transcripts "ENSMUST00000162897"; exonic_part_number "001"; gene_id "ENSMUSG00000051951"
1       dexseq_prepare_annotation.py    exonic_part     3206523 3207317 .       -       .       transcripts "ENSMUST00000162897+ENSMUST00000159265"; exonic_part_number "002"; gene_id "ENSMUSG00000051951"
1       dexseq_prepare_annotation.py    exonic_part     3213439 3213608 .       -       .       transcripts "ENSMUST00000159265"; exonic_part_number "003"; gene_id "ENSMUSG00000051951"
1       dexseq_prepare_annotation.py    exonic_part     3213609 3214481 .       -       .       transcripts "ENSMUST00000162897+ENSMUST00000159265"; exonic_part_number "004"; gene_id "ENSMUSG00000051951"
1       dexseq_prepare_annotation.py    exonic_part     3214482 3215632 .       -       .       transcripts "ENSMUST00000070533+ENSMUST00000162897+ENSMUST00000159265"; exonic_part_number "005"; gene_id "ENSMUSG00000051951"

Different errors when using -f bam

$ python dexseq_count.py -p no -s no -f bam accepted_hits.bam dexseq.gtf dexseq.out
Traceback (most recent call last):
  File "dexseq_count.py", line 94, in <module>
    for f in  HTSeq.GFF_Reader( gff_file ):
  File "/usr/local/lib/python2.7/dist-packages/HTSeq/__init__.py", line 207, in __iter__
    strand, frame, attributeStr ) = line.split( "\t", 8 )
ValueError: need more than 2 values to unpack

----------------------------

Hi again, 

Strange, I could not reproduce the error message, this is what I did:

wget ftp://ftp.ensembl.org/pub/release-83/gtf/mus_musculus/Mus_musculus.GRCm38.83.gtf.gz
gunzip Mus_musculus.GRCm38.83.gtf.gz
python /g/huber/users/reyes/Rpcks/branches/DEXSeq/inst/python_scripts/dexseq_prepare_annotation.py -r no Mus_musculus.GRCm38.83.gtf Mus_musculus.GRCm38.83.DEXSeq.gtf

And then in python, I ran the part of the script that uses the GFF reader:

>>> gff_file = "Mus_musculus.GRCm38.83.DEXSeq.gtf"
>>> features = HTSeq.GenomicArrayOfSets( "auto", stranded=True )
>>> for f in  HTSeq.GFF_Reader( gff_file ):
...    if f.type == "exonic_part":
...       f.name = f.attr['gene_id'] + ":" + f.attr['exonic_part_number']
...       features[f.iv] += f
...
>>>

But I did not get an error message. Could you include the code that you are using?
Alejandro