[Biopython-dev] GSoC python variant update

[Peter Cock]

On Wed, Aug 8, 2012 at 2:55 PM, Brad Chapman <chapmanb at 50mail.com> wrote:
>Lenna wrote:
>> * Any advice about circular genomes or strandedness is also welcome.
>
> Circular handling is an unresolved issue in Biopython:
>
> https://redmine.open-bio.org/issues/2578
>
> It's a bit tricky, especially with features that span the origin.
>
> I'd prioritize handling strandedness since you're going to have plenty
> of reverse strand coding sequences. You're mapping not only within the
> coding region but also back to the original sequence on the reverse
> strand. So in your g2c mapping, the original gene goes from
> e1 -> s1 -> e0 -> s0 as you read 5' to 3' across the sequence. The best
> place to get started is to pick a reverse strand gene and then work
> through the mappings, thinking through the orientations. I find drawing
> it out to be the easiest way.

And then think about mixed strand genes, e.g. transpliced tRNA is
a good example - there is a GenBank example in our unit tests.

Peter