Re: AMBER: Is the MD simulation normal

-- 
   Shulin ZhuangDear Prof. Walker Ross,
> One is the initial position of the ligand. How was this determined?
> Is there is a crystal structure for it or did somebody dock it in place?
 
 I use the crystal structure with a resolution of 2 Å 
>Take a close look at any close contacts between the protein and the
ligand perhaps it is too close. Especially when it was protonated. You
may need to minimise the protons first and then heat slowly etc.
 After building the prmtop and inpcrd files with xleap, I use the
command 'checkoverlap' to check the inpcrd file using ptraj module,
and no error message., which may shows that no too close atoms.
> Secondly take a look at any potential hydrogen bonds between the ligand and
> the protein. In you initial structure are things orientated correctly for
> these bonds? 
 
I carefully examine the hydrogen donor and acceptor of the potential
hydrogen bonds involved in the ligand, especially the hydrogen
position. All seems normal.
>
> Thirdly there may be problems with the parameters for the ligand. How did
> you obtain parameters and charges for the ligand? 
I use RESP method to derive the parameters.  For the atom types etc of
the ligand, all are right.
> Finally, is the ligand known to be stable in the binding site? It may be
> that in reality it doesn't bind and what you are seeing is correct...
The ligand has a Ki value of 7  M. The binding may be weak. This
ligand is a competitive inhibitor for my protein and one paper
described it in detail. However, the bingdinf affinity may be weak
judging from the Ki value.
 I think the weak binding of the ligands may partly leads to this problem.
Best regards
shulin
Chemistry Department 
Zhejiang University PRC
shulin.zhuang.gmail.com
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu