[MMTK] positions and PDB for aminoacids & proteins

[Konrad Hinsen]

On Tuesday 27 May 2003 16:47, Konstantinos Makrodimitris wrote:

> >>>configuration[0]
>
> AminoAcidResidue ALA:
>   Atom N at [-1.0529999999999999, 1.3, 0.61399999999999999]
>   Atom CA at [-0.30399999999999999, 0.032000000000000001, 0.746]
>   Atom C at [0.77000000000000002, -0.014, -0.311]
>   Atom O at [1.952, -0.16700000000000001, -0.047]
>   Atom H at [-1.8049999999999999, 1.385, 1.3859999999999999]
>   Atom O at [0.35399999999999998, 0.125, -1.5669999999999999]
>   Atom H at [-1.522, 1.3680000000000001, -0.35799999999999998]
>   Atom H at [0.17599999999999999, 0.012999999999999999, 1.74]

There are several atoms with the same name (H, O), which is the cause of =
the=20
error message. I suspect that your PDB file does not respect the PDB nami=
ng=20
conventions for atoms in proteins.

> Does this mean that the order of atoms in a PDB for alanine is not
> compatible please?

The order is not relevant, within a residue the atoms are identified by n=
ame.

> I tried also:
> >>>a1 =3D Residue('alanine')
> >>>a1.atoms[0]
>
> Atom alanine.peptide.C
>
> >>>a1.atoms[0].position()

The result is None (not printed by Python), meaning that the position is=20
undefined. Which it is, of course, because you didn't provide a conformat=
ion.

> So it gives positions for peptide atoms but not for sidechains!! Why?

The positions are assigned as long as the PDB names are compatible with t=
hose=20
in the MMTK database (the details are a bit messy, because MMTK makes a b=
ig=20
effort to deal with many non-standard naming schemes). If your PDB file h=
as=20
the right names for the peptide group, but not for the sidechains, you ge=
t=20
the behaviour that you observe.

> Also how does the following function work ?
> I mean if I use Scientific function is there a way to extract positions=
 and
> coordinates:
> >>>from Scientific.IO.PDB import *
> >>>configuration =3D AminoAcidResidue('alanine.pdb')

Please check the documentation. The argument for AminoAcidResidue is the =
name=20
of the residue, not a PDB file name.
--=20
-------------------------------------------------------------------------=
------
Konrad Hinsen                            | E-Mail: hinsen@cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------=
------