[gmx-users] Free Energy Change of Dihedral Restraints.

Drew andrew.pearce13 at imperial.ac.uk
Wed Jan 23 13:14:16 CET 2019


Hi all,

I'm trying to perform a free energy calculation to understand the impact 
of some dihedral restraints I am using to ensure planarisation of the 
backbone of some oligomers I am working on. I've previously done a fair 
few Free Energy of Solvation calculations and I had noticed that there 
exists the option of using restraint-lambdas however I'm having trouble 
getting it to work.

I had assumed I could include the line in the MDP file:

restraint_lambdas        = 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 
0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

and it would modify the dihedral_restraints element of the topology file.

[ dihedral_restraints ]
; ai   aj    ak    al  type  label  phi  dphi  kfac  power
    15   10   70    75      1    180    0     1000
    11   10   70    71      1    180    0     1000
    80   79   139   140     1    180    0     1000
    84   79   139   144     1    180    0     1000
   153  148   208   213     1    180    0     1000
   149  148   208   209     1    180    0     1000
   218  217   277   278     1    180    0     1000
   222  217   277   282     1    180    0     1000
   291  286   346   351     1    180    0     1000
   287  286   346   347     1    180    0     1000
   356  355   415   416     1    180    0     1000
   360  355   415   420     1    180    0     1000
   429  424   484   489     1    180    0     1000
   425  424   484   485     1    180    0     1000

So that as we move from Lambda=0 to 1 all the restraints go from off to on.

When I run the calculation I get the Warning:/"The lambda=0 and lambda=1 
states for coupling are identical"./

I've scoured the gmx-users archive and I think I need to use the pull 
commands in addition to the free-energy ones (pull-coord1-k and 
pull-coord1-kB) to set what state 0 and state 1 is but I can't seem to 
get that to work. The other option I can think of is to change kfac 
myself for each of those Lambdas and the do the thermodynamic 
integration myself but I'd feel a bit lost about how to do that.

Any help or advice would be much appreciated!

Thanks in advance,

Drew



More information about the gromacs.org_gmx-users mailing list