[gmx-users] Comb. Rule 2 LJ Params. Being Stored in .tpr as Rule 1 and Vice Versa?

[Barnett, James W]

I have combination rule set as 2 in my [ defaults ] section. However, when I do gmx grompp and then gmx dump on the .tpr file, the values indicate they are being stored as combination 1. My understanding from the manual section 5.3.2 is that for combination rule 2 the last two values on an [ atomtypes ] line are sigma and epsilon. But, the values stored in the .tpr file are c6 (4*eps*sig^6) and c12 (4*eps*sig^12) according to gmx dump.


Here is an example line from my topology and the corresponding line from gmx dump for both comb. rule 2 and 1:


[ atomtypes ]
; name      bond_type    mass       charge   ptype   sigma        epsilon
  CH3_C     CH3          15.03500   0.000    A       3.75000e-01  8.14817e-01; [CH3]-CHX


[ atoms ]
;   nr       type  resnr residue  atom   cgnr     charge       mass  typeB    chargeB      massB
     1      CH3_C      1    NPA     C1      1          0     15.035   ; qtot 0


Combination rule 2 yields in the .tpr file (should be epsilon and sigma, but are actually c6 and c12):

functype[0]=LJ_SR, c6= 9.06374492e-03, c12= 2.52054979e-05


Combination rule 1 yields in the .tpr file (should be c6 and c12, but are actually epsilon and sigma):

functype[0]=LJ_SR, c6= 3.75000000e-01, c12= 8.14817011e-01


I am just concerned that this is behavior is opposite of what I expected. Perhaps I am misunderstanding how GROMACS stores these values in the tpr file or the section in the manual? Can someone point me in the right direction?


James "Wes" Barnett

Ph.D. Candidate

Chemical and Biomolecular Engineering


Tulane University

Boggs Center for Energy and Biotechnology, Room 341-B