[Biopython-dev] Ndb

[Cayte]

After checking further, I found that the codes were pdb hetero codes.  But I
still have the problem of representing mixed alphabets.  Seq seems to assume
equal sized elements.  If all the letters are mashed together you won't be
able to tell 3 letter codes from 1 letter codes.  I'd appreciate your ideas
on how to handle pdb mixins.

One possibility is a hetero class with the pdb code plus its location on the
sequence.

Also I'm not sure if the mixin is treated as a residue

a doc t
  I t the first or third residue?

                                                      Cayte

----- Original Message -----
From: "Cayte" <katel at worldpath.net>
To: <biopython-dev at biopython.org>
Sent: Thursday, February 07, 2002 4:06 PM
Subject: [Biopython-dev] Ndb


> Ndb has a lot of sequences with more than 2 strands.  Some of the
sequences
> have 3 letter codes embedded in the ACGT alphabet.  I wonder if these are
> annotation.  One of these is YRR. After searching the internet I found
that
> YRR and YRY are sometimes used to annotate parallel or antiparallel.
There
> were other codes like D1T but I couldn't find any explanation.  Should I
> assume these are annotation? Does anyone know what they are?
>
>                                                                Cayte
>
> _______________________________________________