search

tdudgeon / simple-simulate-complex

Simple protein-ligand complex simulation with OpenMM
73 stars 16 forks source link

Separated complex in simulation results #6

Open ivandon15 opened 5 months ago

ivandon15 commented 5 months ago

Hi,

Have you ever encountered a situation in your simulation results where the protein's position remains roughly unchanged, but the ligand jumps from one side to the other (seemingly influenced by boundary conditions)?

image

Besides that, there have also been instances where the ligand and the protein are very far apart, appearing to be unrelated to each other. (Figures depicted as below)

image image

This really confuses me, and I also tried to set enforcePeriodicBox=True when outputting the traj.

Thanks for your contributions! It really helps me a lot.

tdudgeon commented 5 months ago

Yes, this looks like a periodic boundary issue. Did you try using the analyse.py script to re-image the trajectory?

ivandon15 commented 5 months ago

Thank you for replying! Yes, I do use analyse.py to reimage. And as mentioned before, I also tried enforcePeriodicBox=True, but it seems that none of them work. Besides, I think probably re-imaging is only changing the position of the entire complex but remain the relative position unchanged, right?