- Posts: 3
self-assembling bilayers from single lipid .gro
- htranz
- Topic Author
- Offline
- Fresh Boarder
Less
More
10 years 7 months ago #2088
by htranz
self-assembling bilayers from single lipid .gro was created by htranz
Hi Martini gurus,
I am a new user with a little MD background working on self-assembling/simulating my own bilayer systems. Using the DSPC tutorial as a template I have been working to simulate systems with 512 lipids (combinations of DPPC and DOPC) in water. I have manually edited the dspc_single.gro file from the tutorial to generate a dppc_single.gro and dopc_single.gro by simply adding/deleting residues and renaming to match the different lipids, but left the coordinates exactly the same from dspc_single.gro.
Is this an appropriate method to obtain accurate self-assembled bilayer simulations? I am worried that the system energy minimization may be inaccurately represented by using the dspc coordinates, is there a better way to generate single lipid .gro files?
Thank you very much for any help!
Holden
I am a new user with a little MD background working on self-assembling/simulating my own bilayer systems. Using the DSPC tutorial as a template I have been working to simulate systems with 512 lipids (combinations of DPPC and DOPC) in water. I have manually edited the dspc_single.gro file from the tutorial to generate a dppc_single.gro and dopc_single.gro by simply adding/deleting residues and renaming to match the different lipids, but left the coordinates exactly the same from dspc_single.gro.
Is this an appropriate method to obtain accurate self-assembled bilayer simulations? I am worried that the system energy minimization may be inaccurately represented by using the dspc coordinates, is there a better way to generate single lipid .gro files?
Thank you very much for any help!
Holden
Please Log in or Create an account to join the conversation.
- mnmelo
- Offline
- Admin
Less
More
- Posts: 89
10 years 7 months ago #2096
by mnmelo
Replied by mnmelo on topic self-assembling bilayers from single lipid .gro
Hi,
The Martini model is quite robust in that respect, so you should have no problem in using DSPC coordinates as a starting point -- even more so if doing a self-assembly.
Just a quick tip: if you're using GROMACS you can get away without having to actually rename atoms in the .gro file. grompp will complain about name mismatches but you can ignore that in this case.
The Martini model is quite robust in that respect, so you should have no problem in using DSPC coordinates as a starting point -- even more so if doing a self-assembly.
Just a quick tip: if you're using GROMACS you can get away without having to actually rename atoms in the .gro file. grompp will complain about name mismatches but you can ignore that in this case.
Please Log in or Create an account to join the conversation.
- Clement
- Offline
- Admin
Less
More
- Posts: 211
10 years 7 months ago #2097
by Clement
Replied by Clement on topic self-assembling bilayers from single lipid .gro
If I understand well your question, you changed on the fly the names of the beads of your lipids from DSPC to match DPPC/DOPC, (and add/remove beads if necessary) and wonder if the conformation is usable as it is?
The answer is yes, the minimization will fix that; the Martini force field is flexible enough to allow this kind of... tweaks.
The answer is yes, the minimization will fix that; the Martini force field is flexible enough to allow this kind of... tweaks.
Please Log in or Create an account to join the conversation.
- htranz
- Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 3
10 years 7 months ago #2113
by htranz
Replied by htranz on topic self-assembling bilayers from single lipid .gro
Thank you very much for your responses. Clement, yes I did just modify the .gro file as you said, glad to hear the molecules adjust acccordingly. On a somewhat related note, I've tried making a Cholesterol single molecule .gro file by editing the dppc_chol.gro file available on the download page. My intention is to insert Chol molecules into a lipid bilayer, however on attempting to self assemble a 50/50 DPPC/Chol system I get some LINCS warnings early on and then the mdrun terminates with the following error:
[leapfrog:14187] *** Process received signal ***
[leapfrog:14187] Signal: Segmentation fault (11)
[leapfrog:14187] Signal code: Address not mapped (1)
[leapfrog:14187] Failing at address: 0x1bd6a390
[leapfrog:14187] [ 0] /lib64/libpthread.so.0 [0x3bf620eca0]
[leapfrog:14187] [ 1] /opt/gromacs/4.5.5-gcc470-ompi145/lib/libgmx_mpi.so.6 [0x2b3762b7ba5b]
[leapfrog:14187] *** End of error message ***
Segmentation fault
Any idea what might be causing this? Thanks in advance for any help.
[leapfrog:14187] *** Process received signal ***
[leapfrog:14187] Signal: Segmentation fault (11)
[leapfrog:14187] Signal code: Address not mapped (1)
[leapfrog:14187] Failing at address: 0x1bd6a390
[leapfrog:14187] [ 0] /lib64/libpthread.so.0 [0x3bf620eca0]
[leapfrog:14187] [ 1] /opt/gromacs/4.5.5-gcc470-ompi145/lib/libgmx_mpi.so.6 [0x2b3762b7ba5b]
[leapfrog:14187] *** End of error message ***
Segmentation fault
Any idea what might be causing this? Thanks in advance for any help.
Please Log in or Create an account to join the conversation.
- Clement
- Offline
- Admin
Less
More
- Posts: 211
10 years 7 months ago #2127
by Clement
Replied by Clement on topic self-assembling bilayers from single lipid .gro
Mmmmh... I think your problem comes from the the structure of the cholesterol itself, which is extremely constrained and quite unstable with a 40fs timestep. We usually simulate it using a dt of 30fs. I think that's the solution. We might have other models available for cholesterol, but not yet well tested or published; I'll forward your message to the people working on it.
Please Log in or Create an account to join the conversation.
- htranz
- Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 3
10 years 7 months ago #2134
by htranz
Replied by htranz on topic self-assembling bilayers from single lipid .gro
I tried rerunning the simulation with 30fs time step and increased system box size, which seems to have helped but it still terminated with the same error after about 50ns...
Please Log in or Create an account to join the conversation.
Time to create page: 0.097 seconds