[Biopython-dev] Weekly Update for Codon Alignment GSoC project

[Zheng Ruan]

Thanks Eric,

It will not be difficult to allow more than 3 nucleotides forward
frameshift. I will include this into my plan. For the backward frameshift
support, more nucleotides support is desirable but difficult. Some function
to help construct codon alignment with other software will be considered.

Best,
Ruan


On Tue, Jul 30, 2013 at 12:52 AM, Eric Talevich <eric.talevich at gmail.com>wrote:

> On Sun, Jul 21, 2013 at 7:55 PM, Zheng Ruan <zruan1991 at gmail.com> wrote:
>
>> Hi,
>>
>> I post an update for the project last week in my blog<http://zruanweb.com/> as
>> well as my plan next week. As the midterm evaluation deadline is
>> approaching, I also include this into it. Thanks for your comments and
>> suggestions.
>>
>> Best,
>> Zheng Ruan
>>
>
> Hey Zheng,
>
> Great progress so far. Your implementation of codon alignment up to this
> point looks more than adequate to me. A couple thoughts:
>
> - Can your implementation detect inserts (forward frameshifts) of more
> than 3 nucleotides, as might be introduced by introns? Or just 1-2 bases?
>
> - Same question for the backward frameshift implementation. One biological
> cause for these backward shifts is ribosomal slippage -- these are usually
> short, e.g. 1 base, so it is not urgent for your implementation to handle
> larger backward shifts if this would be more difficult. (For drastic
> differences between protein and nucleotide sequences, a bioinformatician
> would normally use some variant of BLAST, exonerate, or another local
> alignment tool, rather than expecting this codon alignment algorithm to
> catch and handle every possibility.)
>
> Have a safe trip home!
>
> -Eric
>