[gmx-users] problem calculating PMF with distant constraints

[rzangi at att.net]

> >>
> >>>I am trying to calculate potential of mean force between two atoms (neon, 
> >>
> >>methane) by running multiples simulations with distances that are constrained 
> >>from lambda=0 to lambda=1 and then integrating <dHdl>. However, this doesn't 
> >>seem to give the right curve as compare to calculations obtained with other 
> >>methods. There appears to be a serious discrepancy.
> >>
> >>the dHdl curve is meaningless, only the integral is physcially 
> >>meaningfull. Of course you can compare to other simulations done in an 
> >>identical fashion.
> > 
> > 
> > We compared difference free energies (i.e. the PMF curves supperimposed on 
> each other) with identical settings and interaction parameters.
> > 
> There are many hidden parameters, not only in gromacs but also in other 
> programs, like combination rules.
> By the way you still don't say how you compute it.

We used the same combination rule as the reference and also used the second type of combination rule.

Below is one of the topology files. Similar results we got also with SPC, SPCE and tip5p, and also with different LJ particles.

The locations of the minimums of the PMF are good but the depth at contact is higher than the state at large distances, where is should be lower.

> 
> 
> > 
> > 
> >>>In particular when the constraint distance was bigger than the cut-off 
> >>
> >>interaction (so the two atoms don't `see' each other) and bigger than the 
> range 
> >>of the solvent induced interactions, the constraint force was not zero but 
> >>replusive. This can be related to what was mentioned in the literature before 
> >>[Chem. Phys. Lett. 196 297 (1992), JCP 100 9129(1994)] as a `dynamic' 
> >>contribution to the free energy when the system allowed to rotate (the 
> >>centripetal force). Thus, even when there are no forces on the atoms, but ones 
> >>allow distance constraint to rotate the atoms will experience a force in the 
> >>opposite direction to their center of mass, which, of course, should not be in 
> >>the PMF - is there such correction to this in Gromacs?
> >>
> >>I don't know what to do about this, there definitely is no standard 
> >>correction.
> >>
> >>
> >>>Nevertheless, if one restrains the two atoms from moving along the x and y 
> >>
> >>axis and sets the constraint distance along the z-axis it still doesn't give 
> the 
> >>right answer.
> >>
> >>>Did someone reported these issuses before                                                                             
> >>
> >>?
> >>I'm not completely sure that you used the right method. If you put a 
> >>constraint between the atoms they won't interact at all, unless you set 
> >>nrexcl to zero. An alternative method is to freeze the Neon atoms and 
> >>measure the force on them.
> > 
> > 
> > nrexcl was set to zero and fn. type 2 for the constraint `bond' was used, so 
> they should see each other.
> "should" is not good enough. have you checked the exclusions in the tpr 
> file?

this is the first few lines of the excl groups from the top file (atom 0 and atom 1 are Neon the rest is water):

            excl[0][0..0]={0}
            excl[1][1..1]={1}
            excl[2][2..4]={2, 3, 4}
            excl[3][5..7]={2, 3, 4}
            excl[4][8..10]={2, 3, 4}

> > 
> >>>This was checked with 3.2.1 and 3.1.4 versions, with a few different water
> >>>models, PME and cut-offs for the electrostatic interactions as well.
> >>>
> >>>Is there a simple fix? The group I am working in is quite concerned about 
> >>
> >>this.
> >>If you have a well defined system where you think you are doing exactly 
> >>the same as in a publication, and you still can not reproduce it then 
> >>feel free to post a bug report on bugzilla.gromacs.org. I assume you are 
> >>reasonably sure that your reference values are correct.
> This comment still holds. For your PMF it could e.g. be a single point.
> 
> 
> 
> -- 



#include "ffoplsaa.itp"

[ atomtypes ]
;name   mass      charge   ptype   sigma        epsilon
LJP   20.1797     0.000     A     0.2749        0.295108

 
[ moleculetype ]
; Name            nrexcl
LJG                 0
 
[ atoms ]
;nr  type  resnr residue  atom   cgnr  charge  mass
1    LJP     1     LJG    LJ1     1     0.0    20.1797
2    LJP     1     LJG    LJ2     1     0.0    20.1797

[ constraints ]
; #at.   f.tp  length_A  Length_B
1   2     2     0.220    1.220

 
[ moleculetype ]
; molname       nrexcl
SOL             2
  
[ atoms ]
; id    at type res nr  residu name     at name         cg nr   charge
#ifdef _FF_OPLS
1     opls_111  1       SOL              OW             1       -0.834
2     opls_112  1       SOL             HW1             1        0.417
3     opls_112  1       SOL             HW2             1        0.417
#else
1       OWT3    1       SOL              OW             1       -0.834
2       HW      1       SOL             HW1             1        0.417
3       HW      1       SOL             HW2             1        0.417
#endif
  
#ifdef FLEXIBLE
[ bonds ]
; i     j       funct   length  force.c.
1       2       1       0.09572 502416.0 0.09572        502416.0
1       3       1       0.09572 502416.0 0.09572        502416.0
  
  
[ angles ]
; i     j       k       funct   angle   force.c.
2       1       3       1       104.52  628.02  104.52  628.02
  
#else
[ settles ]
; i     j       funct   length
1       1       0.09572 0.15139
  
[ exclusions ]
1       2       3
2       1       3
3       1       2
#endif

 
[ system ]
; Name
LJG   SOL
 
[ molecules ]
; Compound        #mols
LJG                 1
SOL                1200