GROMACS version:2022
GROMACS modification: Yes/No
Here post your question
I am trying to run the command for CG MD simulation of graphene:
gmx_mpi grompp -f minim.mdp -c system.gro -r system.gro -p topol.top -o em.tpr
but I am getting the error: Atomtype CG2R61 not found
I have generated the structure and obtained forcefield and .itp files from CHARMM-GUI, I do not get how to solve this error. Can someone please help?
Are you sure this is a CG MD? The atomtype you described is an all-atom atom for carbon nanoparticles.
Regarding the error, check the following:
- You are using the force field files obtained from C-GUI. These are usually in the toppar folder.
- topol.top has #include blocks. Verify the path of force field files is correct.
I have run multiple graphene and CNT simulations using input files generated by C-GUI without this error.
I have generated the all-atom pdb structure from charmm-gui which i further want to use in cg simulation. The CG pdb structure I have made using CGBuilder tool in VMD, for which I go the .apr, .top, and .pdb files but I am getting this Atomtype error and also few additional ones. So, if you could kindly suggest how to proceed or resolve this, it would be of great help. Thank You.
Which CG force field do you intend to use? If it is Martini FF, the atomtype (actually, beadtype) will be C1, SC1 or something else based on the AA-to-CG mapping, functional groups, etc. There are many reports on 3-to-1 and 4-to-1 mapping schemes where 3 or 4 carbons are mapped to a single CG bead. I haven’t worked on other CG FF so can’t comment on those.
Yes I am using the Martini FF. I will check if these atomtype works. Thank you
Things are not that simple. The bonded parameters (bonds lengths, bond angles, etc.) will change. So, just replacing atomtype wouldn’t suffice. I would suggest go through the literature troghly for your specific end use. For example, Titov (2010) and Shih (2015) both use a 3-to-1 mapping but SC4 and SC3 beads, respectively. Also look Wallace (2007) which uses 4-to-1 mapping. There is still no consensus which model best describes graphene. But looking at Martini guidelines, I suggest that you go for a 3-to-1 mapping with small beads. There is also a CNP bead described by Luca Monticelli which may be adapted in your study. But note that the arrangement of small beadtypes won’t, typically, be in a hexagon but in triangles.
Noted with thanks. I have been going through and making certain changes but still getting the error of Atomtype 1 not found