[gmx-users] Fatal error: atoms 970 and 973 in charge group 600 are in different energy groups

[Mark Abraham]

Mao-Cai Yan wrote:
> Dear colleague,
>  
> I want to calculate the Coul-SR energy and LJ energy between the 
> receptor and PART of my ligand, using a given XTC file with "mdrun 
> -rerun". I modified "energygrps" in the MDP file, and then type
>  
> grompp -f md_rerun.mdp -c pr_out.gro -p receptor.top -n index.ndx -o 
> rerun.tpr
>  
> Here is the wrong information:
>  
> 
> creating statusfile for 1 node...
>  
> Back Off! I just backed up mdout.mdp to ./#mdout.mdp.1#
> checking input for internal consistency...
> calling /lib/cpp...
> processing topology...
> Generated 1284 of the 1485 non-bonded parameter combinations
> Excluding 3 bonded neighbours for Protein 1
> turning all bonds into constraints...
> Excluding 3 bonded neighbours for AHE 1
> turning all bonds into constraints...
> Excluding 2 bonded neighbours for SOL 8605
> turning all bonds into constraints...
> Excluding 1 bonded neighbours for Cl 1
> turning all bonds into constraints...
> processing coordinates...
> double-checking input for internal consistency...
> Velocities were taken from a Maxwell distribution at 300 K
> renumbering atomtypes...
> converting bonded parameters...
> #     ANGLES:   6004
> #      PDIHS:   2890
> #      IDIHS:   2400
> #       LJ14:   5595
> #     CONSTR:   3078
> #     SETTLE:   17210
> Walking down the molecule graph to make shake-blocks
> initialising group options...
> processing index file...
> Not using any simulated annealing
> Making dummy/rest group for Acceleration containing 26823 elements
> Making dummy/rest group for Freeze containing 26823 elements
> Making dummy/rest group for Energy Mon. containing 25851 elements
> Making dummy/rest group for VCM containing 26823 elements
> Number of degrees of freedom in T-Coupling group Protein is 1882.89
> Number of degrees of freedom in T-Coupling group AHE is 111.99
> Number of degrees of freedom in T-Coupling group SOL is 51627.11
> Number of degrees of freedom in T-Coupling group Cl is 3.00
> Making dummy/rest group for User1 containing 26823 elements
> Making dummy/rest group for User2 containing 26823 elements
> Making dummy/rest group for XTC containing 26823 elements
> Making dummy/rest group for Or. Res. Fit containing 26823 elements
> T-Coupling       has 4 element(s): Protein AHE SOL Cl
> Energy Mon.      has 3 element(s): Protein disaccharide rest
> Acceleration     has 1 element(s): rest
> Freeze           has 1 element(s): rest
> User1            has 1 element(s): rest
> User2            has 1 element(s): rest
> VCM              has 1 element(s): rest
> XTC              has 1 element(s): rest
> Or. Res. Fit     has 1 element(s): rest
> Checking consistency between energy and charge groups...
> Fatal error: atoms 970 and 973 in charge group 600 are in different 
> energy groups
>  
>  
> I am not sure what "charge group 600" means. Atoms 970 and 973 are in my 
> ligand; atom 970 is in the part of ligand that I want to calculate 
> energy (that is, "disaccharide" in "Energy Mon."), but atom 973 is not.

Charge groups are covered in the manual. See there for details. Since
GROMACS evaluates non-bonded energies as sums over charge groups, not
sums over atoms, it is not possible to have an energy group that is not
a union of whole charge groups. You could re-define the charge groups in
the topology (see chapter 5 of the manual) if you were prepared to have
more noise as atoms move in and out of the cutoffs, which might be
acceptable in a mdrun -rerun context.

Still, I'd suggest strongly that you consider what you expect such an
analysis to tell you. Protein-ligand interactions are very unlikely to
be decomposable into pairwise-additive terms that are a function of
subsets of coordinates - especially enzymes.

> I am a jackeroo and want to get your advice. Many thanks!

Somehow, I doubt it! See http://www.jjoz.com.au/jackjill.htm :-)

Mark