[gmx-users] Alcohol Dehydrogenase MD simulation....Reg

[Mark Abraham]

On 23/03/2011 4:38 AM, Kishore wrote:
> Dear All,
>
> I had been in the look out for someone who has a reasonable amount of 
> expertise in dealing with the MD simulation of an Alcohol 
> Dehydrogenase. I have a problem dealing with the NADPH. I was unable 
> to maintain the planarity of the carboxamide of the nicotinamide 
> moiety in the simulation as a result of which important sidechain 
> hydrogen bonds are being lost in the simulation. Is such behaviour 
> expected or is it because of the non quantum mechanical treatment of 
> the NADPH molecule that these kind of situations occur? If anyone has 
> some experience please share it.

Classical MD simulations use improper dihedrals to maintain such 
planarity. You can see this in the amide links in a peptide backbone, 
for example. Each set of four co-planar atoms (three arranged around a 
central one) needs an improper dihedral. So you need H-N-C-H, N-C-O-Cr4 
and C-Cr4-Cr5-Cr3 (where the ring carbons are named Cr and numbered 
clockwise from N) in order for the carboxamide to stay coplanar with the 
aromatic ring