[MMTK] Protein.py, PeptideChain question

[khinsen at cea.fr]

On Apr 11, 2005, at 15:20, Justin Lemkul wrote:

> I have attempted reading in my arbitrary peptide sequence (ACE Glu Pro 
> His Ala
> Gly
> Phe NME) two different ways:
>
> universe.protein = Protein(filename, "no_hydrogens")

The correct call is

universe.protein = Protein(filename, model="no_hydrogens")

> I assumed that there was some incompatibility between the function and 
> this
> short chain, in that it was a short chain, not a full protein with 
> multipled,
> numbered chains.

No. You can use Protein() for a single chain, of any length.

> So, after a bit of reading, I discovered PeptideChain, and decided to 
> give it
> a shot:
>
> universe.protein = PeptideChain(filename)

The PeptideChain class does not accept filename arguments, it wants a 
sequence.

There are two levels at which protein objects can be created in MMTK:

1) Using Protein() with a PDB file. This works for the most common 
cases.

2) By creating first a PDBConfiguration object, and then creating 
PeptideChain
    objects and finally a Protein object. This is bit more work, but 
provides
    more flexibility at each step. See Examples/Proteins/analysis.py for
    an illustration, and the manual for all the details.

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Laboratoire Léon Brillouin, CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: khinsen at cea.fr
---------------------------------------------------------------------