Hello everyone,
I am simulating a protein-water system in NVT ensemble using GROMACS. Prior to that I have energy minimised the system using steepest descent and conjugate gradient methods. After that the system was equilibrated in NVT and NPT ensemble for 10 ns (first with position restrain and then without it). The final production run was carried out in NVT for 25 ns using the last configuration of NPT equilibration.
After the trajectory is generated I looked at the average pressure and found that sometimes it is ~40-50 bar; in some other run with different starting configuration it is very low (below ~0.05 bar even negative!)
I want to know if this is reasonable or something is going wrong? What should I check to find out if my simulation is going in the wrong direction?
Please drop your valuable comments and suggestions.
Hi Sayantan, it could be related to the thermostat that you used to modulate the temperature. What is the thermostat that you used?
Since you are running your final simulation based on NVT ensemble, it is expected that the pressure is not going to be tightly controlled. As I know, the sensitivity of pressure fluctuation is also related to the size of the 3D simulation box as well. However, in order to verify the accuracy of your protein-water MD simulation, you should at least run your simulation with the water only and see whether density matches with the reported values in the literature. I suppose that you are using TIP3P water model and it should fit quite well with the actual water density at 1atm and 25C.
Hi Sayantan! The pressures of your simulation seem quite OK to me. As mentioned above, the small size of the simulation box (compared to a macroscopic system) makes pressure fluctuations very large (often of the order of a few hundred bars), and this even when using a barostat in a NPT simulation. Therefore I think that getting pressures of the order of +/-100 bars is quite normal.
- Badges
- Science topic