normal Protein simulation using the ElNeDyn force field

  • ruoxu
  • ruoxu's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
9 years 10 months ago #3830 by ruoxu
Hi All:

I am trying to do simulations of membrane proteins by using the ElNeDyn method for the protein, when I following the tutorial on the Martini website, I encountered two problems,

1, When i use the following command to transfer protein to CG model, i get an error:

COMMAND: ./martinize.py -f 1UBQ.pdb -o system.top -x cg_1UBQ.pdb -dssp ./dssp-2.0.4-linux-amd64 -p backbone -ff matrini22 -elnedyn -ef 500 -eu 0.9

ERROR: Traceback (most recent call last):
File "./martinize.py", line 5062, in <module>
options = option_parser(args,options,lists,version)
File "./martinize.py", line 339, in option_parser
options[ar].setvalue([args.pop(0) for i in range(options[ar].num)])
KeyError: '-elnedyn

However, this problem could be solved by using another command, which is:

./martinize.py -f 1UBQ.pdb -o system.top -x cg_1UBQ.pdb -dssp ./dssp-2.0.4-linux-amd64 -p backbone -ff elnedyn -ef 500 -eu 0.9

Is the second command appropriate method to generate the CG model of protein?
Another thing is that, i need to include the itp file for the force field in the top file, does any martini FF compatible with the ElNeDyn method?

2, when I try to generate the .tpr file by the grompp command, I get the following warning:

WARNING 1 : The bond in molecule-type Protein_A between atoms 1 BB and 3 BB has an estimated oscillational period of 9.7e-02 ps, which is less than 5 times the time step of 3.0e-02 ps. Maybe you forgot to change the constraints mdp option. I tried to modify the constraint conditions in the mdp file, but I still get this warning. If I just ignore this warning, my simulation will stop very quickly because of LINCS WARNINGS. But I couldn't use small integration step because too small step is not recommodnated for CG simulation. Could any one help me to figure out where is wrong? How could I solve this problem and run CG simulation by the ElNeDyn method? Thanks very much for your help. I really look forward to hearing any advices from you. With my best regards, RuoXu Gu[file system.top, line 14]:
The bond in molecule-type Protein_A between atoms 1 BB and 3 BB has an
estimated oscillational period of 9.7e-02 ps, which is less than 5 times
the time step of 3.0e-02 ps.
Maybe you forgot to change the constraints mdp option.

I tried to modify the constraint conditions in the mdp file, but I still get this warning.

If I just ignore this warning, my simulation will stop very quickly because of LINCS WARNINGS. But I couldn't use small integration step because too small step is not recommodnated for CG simulation.

Could any one help me to figure out where is wrong? How could I solve this problem and run CG simulation by the ElNeDyn method?

Thanks very much for your help. I really look forward to hearing any advices from you.

With my best regards,
RuoXu Gu

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

More
9 years 10 months ago - 9 years 10 months ago #3833 by Clement
In the first martinize.py command you try to specify too many force fields at a time. ElNeDyn is a force field in itself, and as such the -elnedyn flag is enough to enable it. No -ff and/or -ef 500 -eu 0.9 are needed.

The warning is not applicable to CG; you can ignore it (-maxwarn 1).
ElNeDyn is applying a lot of constraints on the protein backbone; it usually takes some time to relax the structure (no clear number here, it can vary depending on your protein size, shape, etc.). And since distances for bonded parameters are extracted from the initial atomistic conformation, sometime a quick minimization and optimization of the atomistic structure can help the equilibration process. And even if you're right when saying a too small time step is not "recommended" for CG simulations - Martini was parameterized for a ~20 to ~40fs timestep in general - it just means that you shouldn't use it for a production run; it's however highly recommended to equilibrate your system... If problems persist you can also put restraints on your backbone to relax the sidechains during a short run, then release it during another equilibration step (with a smaller time step), and slowly increase the time step in sequential equilibration runs.

In conclusion: -elnedyn flag in martinize.py, and a long equilibration process with a smaller time step and/or sequential releasing of the protein parts.
Last edit: 9 years 10 months ago by Clement.

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

Time to create page: 0.091 seconds