[gmx-users] Can I use slipids.ff + charmm36.ff to create a system?

[Carlos Navarro]

Dear gmx users,

This is the third time I’m sending this email, but for some reason is not
been accepted.

I’m trying to set up a system consisting in a protein embebed into a
bilayer. The whole system contains approximately ~400000 atoms, so to speed
up the simulation I'm using virtual sites.

I had no issue by converting my initial structure (protein) into a .gro+vs
And for the lipid coordinates and parameters I’m using the one published in
the following article:
http://pubs.acs.org/doi/pdfplus/10.1021/acs.jctc.5b01202 files here->
https://zenodo.org/record/47649#.WZssiHeGMcg (In the link are the
parameters for POPC, as well as the ffbonded and ffnonbonded for the slipid
forcefield).

Unfortunately, since the lipids posses its own forcefield file (located in
slipids.ff) I don’t how how to include the parameters in the ‘main’
topology file, because if I do add the slipid forcefield in I get the
following error:

Fatal error:

Syntax error - File forcefield.itp, line 5

Last line read:

'1 2 yes 0.5000 0.8333'

Found a second defaults directive.

error, which make sense.
But also If I tried to add only the .itp file of POPCvs to the .top file
gromacs complain about the atomtypes:

ERROR 1 [file POPCvs.itp, line 66]:

  Atomtype MCH3N not found


If I’m not mistaken, I have read some articles were people indeed can
combine different forcefield, but sadly I don’t know how to do that.
If this is imposible to do, does someone know vsite parameters for POPC
that can be use in charmm36?
Hope someone can help me.
Best regards,
Carlos