I've run a simulation for an apoprotein using fourierspacing = 0.16, and I want to run a protein-ligand complex simulation (all using Amber ff). I found that it's suggested to set the value of fourierspacing to 0.125, if I did so, will it affect the results of the simulation much when I reach the point of comparing the apoprotein and the protein-ligand complex simulation? or should I keep it to 0.16?
It is better to use the same fourierspacing for both systems. At least three trajectories should be generated for each system, so now you have 1 of 6 done rather than 2 of 6, not that big difference in the grand scheme of things.
Thank you @Martin Klvana for your answer, but can you elaborate more what 3 trajectories I should get and how is it calculated 1/6 or 2/6?
Sorry I'm new to Molecular dynamics!
At least three MD trajectories for each system, i.e., at least six MD trajectories for two systems by varying the initial conditions (typically random seed for generating initial velocities). My understanding is that so far you have done one MD simulation for each system (so you have 2 out of 6 done) but you will have to discard one or the other because they were not obtained using the same parameters. So, effectively, you have done 1 out of 6 MD simulations.
Thanks Martin Klvana!
So I should repeat the simulation of each system 2 more times with the more accurate fourierspacing of 0.125 (keeping that one simulation of each system of 0.16 fourierspacing) from the same starting conditions changing only the random seed for generating initial velocities in the NVT equilibration step, and that is different from using enhanced sampling method, am I right?
- Badges
- Science topic