normal Simulation of Proline mutant in membrane blows up

  • simon
  • simon's Avatar Topic Author
  • Visitor
6 years 11 months ago #7229 by simon
Hi

I try to simulate a protein (100 proteins in membrane, starting structures sampled from AA simulation) where a glycine was exchanged with a proline in POPC and POPE/POPG membranes, but after several microseconds (20 fs timestep a bit better than 30 fs) the simulation blows up (domain decomposition or LINCS warnings ). The WT and a leucin mutant run without any problems. I am using martini 2.2 and gromacs5.1 for running and martinize.py and Insane to build the system.

Are there any known issues with proline which can cause my problems or has maybe someone else an Idea why my simulation blows up.

Thanks for the help in advance
Simon

Please Log in or Create an account to join the conversation.

More
6 years 11 months ago #7230 by peterkroon
Replied by peterkroon on topic Simulation of Proline mutant in membrane blows up
Hi,

a proline can cause large changes in structure. Are you sure your starting structure is well minimized/equilibrated?
Otherwise, consider running a short (!) minimization of your proline mutant in vacuum and water.

Please Log in or Create an account to join the conversation.

  • simon
  • simon's Avatar Topic Author
  • Visitor
6 years 11 months ago #7231 by simon
It should be well equilibrated. If I look at ptotential energy and maximum force after minimization it is even better equilibrated than the other simulations which are running.

Please Log in or Create an account to join the conversation.

More
6 years 11 months ago #7232 by peterkroon
Replied by peterkroon on topic Simulation of Proline mutant in membrane blows up
Those absolute numbers are meaningless (except maybe for the maximum force).
Have you redetermined the secondary structure of your protein after the mutation, and how do you enforce that in your Martini model? The options are either an elasic network, or extended dihedrals.

Please Log in or Create an account to join the conversation.

  • simon
  • simon's Avatar Topic Author
  • Visitor
6 years 11 months ago #7233 by simon
My starting structures were sampled from a previous AA simulations and then Coars-Grained with the martinize script without further modifications. The DSSP output from the martinize script shows that the proline as an helix residue. Could this cause the problems that proline was acounted as an helix residue which should not be correct.

Could it help to give martinize a other secondary structure (-ss option) ?

Please Log in or Create an account to join the conversation.

More
6 years 11 months ago #7234 by peterkroon
Replied by peterkroon on topic Simulation of Proline mutant in membrane blows up
That's worth a try at least.
What do you get if you run dssp on your AA trajectory (instead of a frame)?

Please Log in or Create an account to join the conversation.

Time to create page: 0.102 seconds