Martin Klvana , The atomtype CH3 was defined as you suggested, which was coded differently in complex.gro file. But also the same error occurred, kindly explain.
i don't use Gromacs, but i assume that for the atom type (CH3) to be recognized the only thing that needs to be done is to define the atom type in the force field file.
A google search will be more helpful. The error indicates the atom CH3 is not defined/identified when compiler map the input files with forcefield parameters. If it is a new molecule you need to make sufficient files to include in ff. If it is biomolecules like proteins, your input files are not in standard form as per GROMACS.
PRODRG server are not good enough for parameterization of OPLS>
@Nikhil, Sir i have added CH3 atoms to the necessary input files, but i am not sure it is correct or not (not sure in which ff file i have to add). And as you mentioned, i have obtained the topology file (.top) and .gro of ligand from PRODRG server. Kindly help.
PRODRG produces (bad) topologies for GROMOS force fields. Do not attempt to mix and match force fields - the resulting physical model is completely invalid, even if you manage to hack the force field files to "make it work." Use the LigParGen server for OPLS topologies.