Pass potential energy components from Gromacs to Plumed

[PattaWapee]

I would like to learn how to construct collective variables that operate on various components of the potential energy function, for instance a protein-ligand interaction energy computed in Gromacs. I would like to find out what is the best way to implement such CV, for instance whether it may be possible to pass components from Gromacs to Plumed, or if energies and forces should be recomputed in Plumed, and if so, what should be the best way to access the required parameters.

[gtribello]

Hello Thanks for your interest in the meeting and for making a suggestion. As you yourself suggest there are two ways that one might imagine doing these things. Either you implement a new CV in plumed or you pass more data from gromacs to plumed. If it is OK with you I would rather opt for the first of these options as I know the insides of plumed but I don't know the insides of gromacs. I think we can what you ask by exploiting the multicolvar functionality. You might thus want to have a look at those bits of the code before you get to the meeting.

Thanks again Gareth