[MMTK] bad lysine protonation

Konrad Hinsen hinsen at cnrs-orleans.fr
Fri Aug 29 07:44:34 UTC 2008


On 28.08.2008, at 02:39, Robert Jorissen wrote:

> I have tried out one of the test scripts from the MMTK web site to  
> read in a protein structure, add hydrogen atoms and then write out  
> a PDB file. However, the lysine sidechain seems to be neutral  
> rather than charged - two hydrogen atoms are added to an apparently  
> sp2 sidechain nitrogen. I also noticed this on another protein that  
> I was working with, so it is not restricted to this particular PDB  
> file.

When reading a PDB file, MMTK has to choose the protonation state of  
the various residues. It does so by choosing the one that is closest  
in number of protons to the one given in the PDB file. For PDB files  
used in molecular simulations, or for those coming from NMR, this is  
probably the most reasonable behaviour.

For hydrogenless PDB files, such as the ones typically obtained from  
crystallography, this leads to the choice of the least protonated  
variant, which is often not the good choice. If anyone has a better  
idea how to deal with this case, I'd like to see it.

As a quick fix to your problem, try the following script. It  
deactivates the identification of the protonation state, and you will  
get the most probable protonation state at pH 7 for each amino acid.

Konrad.


from MMTK import *
from MMTK.PDB import PDBConfiguration
from MMTK.Proteins import PeptideChain, Protein

class MyPDBConfiguration(PDBConfiguration):
     def identifyProtonation(self):
         return

configuration = MyPDBConfiguration('1B2F.pdb')
configuration.peptide_chains[1][28].name = 'LYS'

chains = configuration.createPeptideChains()

protein = Protein(chains)

protein.writeToFile('protein_with_h.pdb')

--
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Moléculaire, CNRS Orléans
Synchrotron Soleil - Division Expériences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: hinsen at cnrs-orleans.fr
---------------------------------------------------------------------






More information about the mmtk mailing list