[gmx-users] Energy minimization, of peptide-ligand, tearing peptide apart

[Richard Kerr]

Hi everyone

I am currently trying to perform a peptide-ligand (~4000 and 110 Da
respectively) simulated annealing study using gromacs (5.1.2) with Amber
force fields.  Under the initial energy minimization step (minim.mdp and
nvt.mdp) I am finding that the step fails.  It then kicks out pdb files,
which when opened depicts a ligand that has 'exploded' with the peptide
fragmented into a number of pieces.

Under the same conditions, I can get the peptide to run with a ligand about
twice the size of the above, which follows through to completion of all
annealing steps to gain output pdb files.  Accounting for a peptide only
system, the peptide will also follow through to completion.

Is anyone able to suggest how to optimize the minim/nvt files, or make
sense of the log file.

I have linked to the minim, nvt and output log files below.
minim
https://www.dropbox.com/s/80wwe8n2ncpg77t/minim.mdp?dl=0

nvt
https://www.dropbox.com/s/qruyanl8vg1a3no/nvt.mdp?dl=0

Log details (md_temp.e20786725)
https://www.dropbox.com/s/0fnjm05ub85yiun/md_temp_e20786725.txt?dl=0

Any help would be greatly appreciated.

Richard