Hi Adam,
well, there is this little gem of a semi-global DNA Protein Aligner:
http://hackage.haskell.org/package/DnaProteinAlignment
As a bonus, the algorithm is explicitly designed to handle frameshifts.
There are some weird species out there...
It's not exactly what you need, but on the other hand it is not too hard
to write these kinds of algorithms, either.
caveat emptor:
- I have a high opinion of the author of this package ;-)
- a new version of the whole framework is being prepared; but it comes
with a transition to using 'stack' and many upgrades -- and hence
takes a while
Viele Gruesse,
Christian
* Adam Sjøgren <a...@koldfront.dk> [01.11.2015 17:46]:
> Christian Höner writes:
>
> > no, but after skimming the paper it doesn't look that complicated to
> > implement. There is also a bunch of related work on how to reduce the
> > work involved in global MSA, i.e. on bounding the maximal deviation from
> > the main diagonal...
>
> > depends on what you want to actually do with the implementation.
>
> What I am really looking for is making DNA-DNA alignments that take
> reading frame/gene location into account - when comparing sequencing
> data to expected sequence, after introducing mutations.
>
> E.g. when 3 bases are deleted, the alignment would tend to remove an
> entire codon, rather than 2 bases from one codon and 1 base from another
> codon.
>
>
> Best regards,
>
> Adam
>
> --
> "Perl is a Shinto shrine." Adam Sjøgren
> a...@koldfront.dk
