[gmx-users] alchemical thermodynamic integration in gromacs 5.0 - bonded lambdas

[Oliwia Maria Szklarczyk]

Dear All,

I would like to run a free energy calculation using thermodynamic integration. The alchemical process I am simulating is a mutation of a alanine to a leucine in a protein. So I am creating new atoms and interactions. I changed the topology file such that new types of atoms, bonds, angles and dihedral angles are written at the end of the line for each atom that is mutated or that is appearing, in respective directive blocks. For atoms that are appearing I create them from dummy atoms. 

My question here is about the mdp file. I have vdw-lambdas and coul-lambdas vectors defined from 0 to 1 with delta_lambda = 0.05, so 21 lambda points. Do I also need to define the bonded-lambdas vector? I have the changes in bond types defined in the topology file, and I see that the bonded energy term changes between state A and B. 

Cheers,
Oliwia