Hi!

I don't know what is your system but your simulation time looks very short -50 ps-, I think you should run longer (at least 500 ps-1 ns, this may be change according to system and initial structure) simulations.

Also you can check the Total Energy plot from log file by VMD (Analysis - NAMD Plot). You should look for RMSD after energy stabilization.

I used -RMSD- to show the stability as well throughout 20 ns. In the earlier times of simulations molecules were unstable and gradually became stable. So I assume you should run it in a longer time, as you were recommended.

Caglanaz Akin How can I stabilise energy ? I just ran 10000 step minimisation. Should I have done something else? what is the best way to stabilise energy ?

Fatemeh Hejazi Amiri even if you run minimization, sometimes at early steps of MD simulation energy is fluctuating. Generally we run equilibration MD before run MD with same parameters to stabilize it.

You can follow these steps:

1. minimize the structure

2. if you minimize the structure @0 K you should heat the system to RT (by the way, people generally run MD simulations at 300-310 K when they are using NAMD, I don't know the reason behind this, you can check it.)

3. after heating, you should run at least 500 ps - 1 ns MD to equilibrate your system

4. after equilibration, you can run MD. When you analyze your MD results, you should use the trajectory from this step.

These steps may be change according to your system, but generally it works. Also you can merge the step 3 and 4 (longer MD simulation).

Hey,

According to your config file, you are performing a simulation with an NPT ensemble. At first, this is really a short time to comment on RMSD.

So basically your system works in 2 steps:

1- You employ Langevin dynamics to balance friction with a random noise at a defined temperature. This thermostat controls the temperature to perform a constant temperature MD.

2- To control the pressure, you are using NH Langevin piston. Pressure is controlled by changing the size of your simulation box (unit cell). Additionally, you are rescaling all atomic coordinates by doing this. This is one of the reasons why your RMSD values increasing fast at the start of simulation.

Here some advices :

Try to fix or apply a constraint to atoms of your protein etc. at the start of the simulation. Then you need to slightly remove these constraints while equilibrating system. However, at a time you should completely remove these constraints because they create a bias on your system and that would be misleading.