[gmx-users] problem in energy minimization

[Mark Abraham]

Abhik Mukhopadhyay wrote:
> Hi Jastin
> My pdb2gmx command line is
> pdb2gmx -ignh -f xx.pdb -o xxout.pdb -p xx.top -water spce
> 
> When I created the topology it did not create a bond between those two 
> residues (between which the part is missing).

The above statement seems inconsistent with the observations below.

>     > I am trying to run a simulation on a xray structure that has 15 residues
>     missing. I have made the topology file accordingly. But after energy
>     minimization, the residues between which the portion is missing, got connected.
>     I tried an energy minimization with all-bonds constraint, but that i guess did
>     not do it properly.

By far the most likely reason for the residues that should be spaced by 
the 15 missing residues to "get connected" was that your topology had an 
interaction defined.

The fact that these residues were missing from the X-ray structure 
doesn't mean you should ignore their existence. Often they can be 
involved in useful chemistry. I would most likely consider using some 
(non-GROMACS) modelling software to "build in" such residues before 
attempting any MD.

There are alternatives which involve capping the pseudo-termini and 
possibly using some kind of restraints, but these will be harder to 
defend to your paper's audience.

Mark