[BioPython] codons and complements
Andrew Dalke
andrew_dalke@hotmail.com
Tue, 12 Oct 1999 13:35:07 MDT
Jeff Chang <jchang@SMI.Stanford.EDU>:
>NCBI has a reference for IUPAC amino acid definitions:
>http://www.ncbi.nlm.nih.gov/collab/FT/aa_abbrevs.html
This is the same as the IUPAC reference, as far as I can tell.
>NCBI has a FASTA format description, where they use '-' for gaps,
I've got no problem having the biopython standard for gaps being
'-' (note, this could be translated to "." or " " for the outside
world; I'm talking about an internal standard.)
>'U' for selenocysteine
Hmm. IUPAC says
| U and O were avoided because U is easily confused with V in
| handwritten material, and O with G, Q, C and D in imperfect
| computer print-outs, and also with zero. J was avoided because
| it is absent from several languages.
I can see having U as the biopython standard name for
selenocysteine. I would rather, however, use IUPAC definitions
as the internal form, but easily (dare I say "transparently" :)
support different alphabets.
>'X' for any amino acid, and '*' for stop codons:
Fine with me, though ditto on the alphabet pedandtry. I don't
expect there will be any problem in the actual use of what I
proposed. All it really does is hide better bookkeepping.
>The peptabiol database has a vocabulary for a bunch of
>non-standard amino acids (usually found in fungi):
>http://www.cryst.bbk.ac.uk/peptaibol
>http://sanda.cryst.bbk.ac.uk/peptaibol/nomenclature.html
Cool! This is exactly what I wanted, since I knew there was
a need to support different alphabets for different uses, and
the mechanism I proposed is able to handle these.
>Using SEC and X for selenocysteine first used in;
>Gunzler WA, Steffens GJ, Grossmann A, Kim S-MA,
>Otting F, Wendel A, and Flohe L. (1984)
>Hoppe-Seylers Z. Physiol. Chem. 365, 195-212.
The IUPAC reference I gave for protein sequences lists the
reference in the addendum.
Andrew Dalke
dalke@acm.org
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com