I have build a very complex membrane system, which I would like to use for MD simulations with GROMACS. I have made some calculations with it and I have come to the idea, that I have to insert a transmembrane protein into it. However, I don't want to build the membrane from the very beginning... Is there any possibility to put it somehow?
Beware in using g_membed; it's not coded in a robust way and was removed from GROMACS because it did not often work as advertised. InflateGRO is another possibility but the script requires modification for accepting multiple lipids. Note that you're introducing a major perturbation to the system by adding the protein, so the pain you may go through to preserve the membrane may end up being a ton more work than just building it anew, because you've got a brand new system that requires extensive equilibration anyway.
Hi,
You can use OPM server to know the orientation of your system on a bilayer membrane. And then use this information to construct your bilayer membrane with charmgui.
https://opm.phar.umich.edu/ppm_server
http://www.charmm-gui.org/?doc=input/membrane.bilayer
Martiniano Bello but that is building the system from the very beginning, which I want to avoid somehow...
Ok, g_membed can be useful for that.
Article g_membed: Efficient Insertion of a Membrane Protein into an ...
Beware in using g_membed; it's not coded in a robust way and was removed from GROMACS because it did not often work as advertised. InflateGRO is another possibility but the script requires modification for accepting multiple lipids. Note that you're introducing a major perturbation to the system by adding the protein, so the pain you may go through to preserve the membrane may end up being a ton more work than just building it anew, because you've got a brand new system that requires extensive equilibration anyway.
If you decide to start from the very beginning you can use CharmmGUI to prepare the protein-membrane system. Then, if you want to use a different force-field of charmm or SLipids, you should translate the topology, etc, but is an easy way to construct it. If not, as Justin says, inflate gro is an option but you should work harder on it....
I am not aware of any other method than ours (mentioned by Wojciech Kopec
above) that does not require any removal of lipids. Just be careful: GROMACS now gives a warning when you try to push the protein in with the default "refcoord_scaling=no" option. However, our approach only works with this option, so you should ignore the warning with the -maxwarn flag. If you decide to use our method and need some help, don't hesitate to send me an email or a private message here.- Badges
- Science topic