[gmx-users] Crush after ~135 ps

[Semen Esilevsky]

Dear GMX users,
Thanks to your help I've set up a simulation of my
protein. It was expected to be 1ns run, however it
crashed after ~135 ps.
The last message in the log was something like this:
"Can't set a water molecule starting from atom ###,
fatal error at bla-bla-bla.c line ###", than I was
asked to reduce the step size or to check the system. 

The system was solwated by ~8000 water molecules in a
rectangular box. PME with the defauls params from the
manual was used.

So far I've found several suspicious points:
1) Too short equilibration of the solvent with
restrained protein
2) Exposure of the hydrophobic areas in the protein
(it  is a membrane protein, but I put it into the box
of water)
3) The system is not neutral (+7)

However, I still can't understand why the simulation
crashed after quite long successful run. 
Any ideas how to fix this problem?

Sincerely,
Semen





__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com