GROMACS version: 2022-3
GROMACS modification: Yes
Hello,
I am attempting to add the Hunter-Sanders pi-pi interaction model in the gromos54a8 force field. As such I’d like dummy atoms with -1/2 charge a distance delta above and below the plane for aromatic rings and +1 coincident with the C (or N) atoms†. I was thinking that virtual sites are the way to go (but not necessarily the ones provided by the -vsite option of pdb2gmx since they are for modeling the rings themselves) because with that I get the benefit of the velocities of the dummy atoms following that of their corresponding real C atom. I was following @jalemkul 's virtual site tutorial but there’s a big difference between my system and his, I have many aromatic residues in mine. I was wondering whether there is a way—maybe by modifying aminoacids.vsd—to allow gmx2pdb to still assemble the itp for me with these virtual sites. Or is it more reasonable to allow gmx2pdb to assemble an itp without virtual sites then have a script go through and inject virtual sites, making sure to reindex the atoms, bonds,…
Or are virtual sites not recommended and I should use, say, DUM atoms and insert my own pi-pi residues into aminoacids.rtp?