error using ipynb to build genomes

[spleonard1]

Trying to follow the Genome db instructions and the initial call to "download_Ensembl_files" is failing with the following error.

>>> gdb.download_Ensembl_files("../snod_genomedb", maxgen=None, names="snodgrassella", complete=False)
Current release of EnsemblBacteria: release-39
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/site-packages/pyparanoid/genomedb.py", line 72, in download_Ensembl_files
    thisline.append(js[feat])
KeyError: 'assembly_level'

Using python2.7. Any ideas?

[ryanmelnyk]

Yeah - unfortunately Ensembl keeps changing the structure of their FTP directories....I fixed it for release-36 but that apparently they've made another change. I get the same error too. I'm about to get on a flight so I won't have time to fix this right now.

Can you give download_Refseq_files() a shot? As of last week, that was still working for me. It should have even more genomes than Ensembl anyways.

[spleonard1]

Yea, thanks! download_Refseq_files() worked great to download the peptide .faa files as .pep.fa

However, it doesn't download the corresponding dna/ or gbk/ files. Whats the best strategy for that?

[ryanmelnyk]

Nice! Dna and gbk files aren't downloaded by default (for storage reasons) and you don't need them to run PyParanoid. But they are very helpful for the downstream stuff. Check out:

gdb.download_dna_files(strains,genomedb)

and

gdb.download_genbank_files(strains,genomedb)

genomedb is the path to your folder containing the dna/ gbk/ and pep/ folders. "strains" is a python list object containing the strain names of interest (see the first couple cells of the CompareStrains notebook).

[spleonard1]

got it, very helpful. Thank you!

[ryanmelnyk]

I fixed download_Ensembl_files() to play nicely with the new Ensembl FTP structure so I will close this issue for now.