[gmx-users] Free Energy Calculations in Gromacs

[HANNIBAL LECTER]

Thank you Professor Shirts.

So now my protein itp file looks like this:
[ atoms ]
;   nr    type  resnr residue  atom  cgnr  charge    mass    typeB
 chargeB  MassB
 1     CT    1    ACE    CH3      1   -0.190264   12.010000      CT_per
-0.120334    12.010000
....
[ bonds ]
;ai  aj func th0 kb th0B kbB
    1     2  1   0.10900  284512.0   0.10900    113804.8
....

[ pairs ]
1     8     1
....

[ angles ]
;ai   aj    ak      funct      th0         kb         th0B          kbB
    2     1     3  1   109.500  292.880   109.500    117.152
.....

[ dihedrals ]
i    j     k    l     func     th0       cth     mult      th0B     cthB
   multB
    2     1     5     6   9    180.0     2.86144  1    180.0    1.14458  1

Now, I have created 6 different lambda values and placed a
well-equilibrated structure and the same topology file (with values
corresp. to lambda=0 and lambda =1) within six directories. The .mdp files
contain the information (init-lambda-state=0,1,..) which lambda state is
it.

Is my approach correct ?

Secondly, when I grompp followed by gmxdump -s *.tpr I find, that there is
no difference in hamiltonian between the two. So, does the scaling take
place during runtime?

Third, since in state B only the hamiltonian is rescaled (the perturbed
entries are defined in ffnonbonded.itp file) do I need to explicitly add
another section [ pairs_nb ]?

Again, I thank you for replying.



On Sat, Apr 20, 2013 at 5:52 PM, Michael Shirts <mrshirts at gmail.com> wrote:

> You have to change atom types.  For example:
>
> [ atomtypes ]
> ;name  bond_type    mass    charge   ptype          sigma      epsilon
> h1            h1         0.0000  0.0000      A      2.47135e-01
>  6.56888e-02
> h1_pert     h1         0.0000  0.0000      A      2.47135e-01  3.56888e-02
> ; perturbed
>
> The mass and charge can be zero, because they will be defined in the [
> atoms ] part
>
> [ atoms ]
> ;   nr    type  resnr residue  atom  cgnr  charge    mass    typeB
>  chargeB  MassB
>     1     h1      1     BLAH    H1a     1     -0.089    1.008    h1_pert
>  -0.030  1.008;  perturbed
>
>
>
> On Sat, Apr 20, 2013 at 4:04 PM, HANNIBAL LECTER <
> hanniballecter13 at gmail.com
> > wrote:
>
> > Hi,
> >
> > I am trying to perform expanded ensemble simulations between 2 states A
> > (lambda=0) and B (lambda =1) with 6 intermediate lambda values.
> >
> > In state B the Hamiltonian is rescaled, such that the epsilons of the vdW
> > interactions, the charges, the bond, angle and dihedral constants are a
> > multiple of the similar terms of State A.
> >
> > It's not quite clear to me (going through the archives of the gmx-users
> > mailing list how to perform these calculations. One way to do this, is to
> > use a single topology file in which the charges and the other terms are
> > specified for both states A and B. However, it is not clear as to how
> > should I scale the epsilons in the topology file. (My atoms are not
> mutated
> > in state B. They are the same atoms with scaled epsilons.)
> >
> > Secondly, since I have the topology files for states A and B, is there a
> > way I could perform the simulations (any particular way in grompp) where
> > both the topology files can be preprocessed designating the end states
> and
> > using the mdp options, the simulations corresponding to the intermediate
> > lambda values performed??
> > --
> > gmx-users mailing list    gmx-users at gromacs.org
> > http://lists.gromacs.org/mailman/listinfo/gmx-users
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> > * Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to gmx-users-request at gromacs.org.
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> --
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>