Complete genomes can be 'complete' or 'chromosome'

[tseemann]

We have a problem with bacterial genomes that sometimes the assembly_summary.txt file says complete and sometimes it says chromosome for finished bacterial genomes.

This is partially to do with the fact that bacteria usually only have 1 chromosome, but partially because they have plasmids too. It's confusing.

If I want both, can I do -l complete -l chromosome to get both?
Or so I run 2 commands with same -o folder?

[kblin]

It seems that chromosome means that there can still be gaps in the assembly, but the gaps need to be of known size. Short reads that likely fill the gaps are frequently also included. I've not seen this in complete records.

In any case, you can always run ncbi-genome-download multiple times with the same output dir and different options to get exactly the set of downloaded files that you want.

Allowing multiple --assembly-level would be nice, as would be allowing multiple --format parameters, and some others. But it also makes the CLI more and more complicated, not to mention the code. :)

[kblin]

I will add this to the documentation. I'm just beginning to think that the features are outgrowing the README file and I need to start doing this properly.

[tseemann]

There is also overlap between taxid and division
eg. using -t 2 (or -T 2 ?) is the same as using bacteria at end.

[tseemann]

I am thinking if --assembly-level or --format contain a "comma" , character then you could convert into a regexp and match using regexps in your filtering logic.

For example --format genbank,fasta,wgs would become ^(genbank|fasta|wgs)$

And --format protein-cds would just be ^(protein-cds)$ etc

I've used this method in my tools before and it works well. You could also use a hash/dict lookup by creating a want_format[ xxx ] = True/False system instead of regexp. This method also has the advantage that if you preset all the valid options to False then you can validate input easily.

[kblin]

That's not quite how the input parsing works, but it shouldn't be too hard to do this.

[kblin]

I think @rhpvorderman's patches take care of this use case, don't you think?

[kblin]

The new 0.2.6 release contains the patch from @rhpvorderman that allows you to use --assembly-level complete,chromosome. That should take care of this use case, right? If not, feel free to reopen this issue.