[Cdk-user] PDBReader and protein handeling

[Martin E. <mar...@fa...>]

Hi,

The way CDK currently reads PDB-files is that different strands of a
multistranded proteins are put as different BioPolymers in a
SetOfMolecules and no connectivity information (implicit or explicit) is
read. As we want to perform forcefield calculations on proteins we need
bonds to be read as well. Hence, I have implemented routines for reading
bonds from (protein) PDB-files (which I can make available). However,
having different strands in different Molecules makes it impossible to
have bonds (e.g. S-S bridges) between strands. These bonds obviously
affect forcefield calculations a great deal and are thus important to
handle. I suggest (as a feature request) that all strands of a protein
are kept in the same BioPolymer object. To handle information about
strands I suggest implementing a new Strand class (much like the Monomer
class). 

>From my colleague Ola I know that people are working on a new
CdkModelAdaptor for Jmol as well as extentions to CML to handle
BioPolymer specific informations. Obviously these things would be
affected by the things I suggest above (which is why I write).

Best regards,

Martin.

-- 
========================================
Martin Eklund
PhD Student
Department of Pharmaceutical Biosciences
Uppsala University, Sweden
Ph: +46-18-4714281
========================================