[Biopython-dev] Bio.GenBank (was: Bio.File)

[Peter Cock]

On Wed, Sep 14, 2011 at 4:34 PM, Michiel de Hoon <mjldehoon at yahoo.com> wrote:
> Hi Peter,
>
> --- On Mon, 9/12/11, Peter Cock <p.j.a.cock at googlemail.com> wrote:
>> How's this?
>> https://github.com/biopython/biopython/commit/ff72037efbae2da6eb6db550aa6d02b883ec1345
>
> The code looks good.

OK.

> About the documentation, at the top of the module you say
> that using Bio.GenBank can be useful for WGS master
> records. That is true, but people with particular interests
> may have other reasons to use Bio.GenBank, and maybe
> WGS master records will not be stored as GenBank files
> in the future. So it may be good to keep the documentation
> a bit more generic, so it's still valid in a few years. But I
> agree that in most cases and for most users, Bio.SeqIO
> is the appropriate module rather than Bio.GenBank.

Please go ahead and try to make it clearer.

> Does Bio.SeqIO still need to use Bio.GenBank's
> FeatureParser? Or can it also use Bio.GenBank.read()
> or Bio.GenBank.parse()?

Yes, and no, respectively. At least as written - I guess
the new read/parse functions could take an optional
argument to control this but I fear that would just be
confusing.

Essentially both are both using the scanner/consumer
model, but one uses the Record producing consumer
and the other the SeqRecord producing consumer.

Peter