[gmx-users] more than one peptide in one simulation box

[Mark Abraham]

On 23/04/10 13:16, shahid nayeem wrote:
> Dear All
> I am trying to study inter peptide interaction fpr which I need to put
> more than one peptide in one simulation box. I did it with genconf
> command but this inserts peptide in a regular ordered manner I want
> these to be in irregular disordered insertion. Even after using genconf

Well that's a difficult and atypical scenario. genconf -shuffle will 
allow you to stack the same peptide in a regular array with random 
rotations of the whole box. Then you can solvate, equilibrate and run MD 
at a high temperature to give yourself a quasi-disordered starting state.

> , I tried to proceed furthe after solvation with spc water. The energy
> minimization (steepest descent) failed to converge even after 5000 steps
> and theirafter position restraint dynamics failed giving segmentation
> fault. Introducing more peptide after generating .gro with -ci -nmol
> gives error showing more than one residue in insert molecule.
> Please help me and write  commands which I should follow.

No, because that's an impossible task. We can't begin to guess the 
reasons for things failing without seeing the actual output (was the EM 
energy large and negative? what was the actual error message  from -ci 
-nmol?).

You should be careful to start with a small test case so that you can 
learn the workflow with a manageable problem. Can you get a single 
peptide to equilibrate? Two stacked peptides? It is best to learn to 
walk before trying to run :-)

Mark