[gmx-users] tutorial for "dual" alchemical transformation in gromacs?

[Sebastian Stolzenberg]

Dear Hannes,

Thank you very much for your answer.

> Practical question: why would you want to do that?
> Your two molecules will drift apart unless parts of the molecules are
> linked together in a "single" topology fashion or or you introduce
> other restraints/constraints.
I am doing some method development, so my plan is to currently fix all
atoms of A and B.

> In practice, you can do that with Gromacs relatively easily.  The
> manual describes how to set up the A and the B state of a molecule.
OK, thanks. As a start, I set up a simple "Cl->Na in a water box" system:

gro file:
    1  Na+  NA+    1   1.434   1.045   1.049
    2  Cl-  CL-    2   1.434   1.045   1.049
    3  WAT    O    3   2.914   1.241   1.997
    3  WAT   H1    4   2.899   1.335   1.984
    3  WAT   H2    5   2.841   1.212   2.052
...

i.e., I transform a "Cl-" into a "Na+" by de-coupling (annihiliating)
the non-bonded interactions of "Cl-" and coupling (letting appear) the
ones for "Na+". This I attempt by defining "dummy" atom types IMX and
IPX for Cl- and Na+, respectively:

top file:
...
[ atomtypes ]
;name   bond_type     mass     charge   ptype   sigma         epsilon
    Amb
 IP       IP          0.00000  0.00000   A     3.32840e-01   1.15897e-02
; 1.87  0.0028
 IM       IM          0.00000  0.00000   A     4.40104e-01   4.18400e-01
; 2.47  0.1000
 IPX      IP          0.00000  0.00000   A     0.00000e+00   0.00000e+00
; 0.00  0.0000
 IMX      IM          0.00000  0.00000   A     0.00000e+00   0.00000e+00
; 0.00  0.0000
...
[ atoms ]
  ; id_    at type res nr  residu name     at name  cg nr  charge   mass
 typeB chargeB   massB
    1       IPX     1          NA+         NA+       1      0
22.9898   IP     0     22.9898
...
[ atoms ]
  ; id_    at type res nr  residu name     at name  cg nr  charge   mass
  typeB chargeB   massB
    1       IM      1         CL-           CL-      1      0
35.45300  IMX    0     35.45300

(Please note that for now, I artificially set the charge of each ion to
zero to avoid total charge differences for different lambda values.
Also, for "Cl->Na", this strategy is certainly more complicated, but in
principle allows for different numbers of atoms between molecule A and B
in the "A->B" transformation.)
Unfortunately, running grompp with the standard run.mdp file from
http://www.gromacs.org/@api/deki/files/196/=fe-tutorial-4.6.pdf
(I omitted "couple-lambda0" and "couple-lambda1"
and set "couple-moltype           = NA+ CL-")
I am puzzled by the following error message:
"The lambda=0 and lambda=1 states for coupling are identical"

Is it possible from this error message to see what I am doing wrong?

Many Thanks,
Sebastian




On 29.11.2015 14:36, hannes.loeffler at stfc.ac.uk wrote:
> Practical question: why would you want to do that?  Your two molecules will drift apart unless parts of the molecules are linked together in a "single" topology fashion or or you introduce other restraints/constraints.
> 
> In practice, you can do that with Gromacs relatively easily.  The manual describes how to set up the A and the B state of a molecule.
> 
> ________________________________________
> From: gromacs.org_gmx-users-bounces at maillist.sys.kth.se [gromacs.org_gmx-users-bounces at maillist.sys.kth.se] on behalf of Sebastian Stolzenberg [s.stolzenberg at fu-berlin.de]
> Sent: 29 November 2015 13:11
> To: gromacs.org_gmx-users at maillist.sys.kth.se
> Subject: [gmx-users] tutorial for "dual" alchemical transformation in   gromacs?
> 
> Dear Gromacs experts,
> 
> I am looking for a Gromacs tutorial to perform a "dual" alchemical
> transformation in a water box, i.e. a mutation or any other alchemical
> transformation, in which:
> the nonbonded interactions between the environment and a molecule A
> disappear as lambda goes from 0 to 1, and
> the nonbonded interactions between the environment and a molecule B
> appears simultaneously.
> Interactions between atoms A and B are to be excluded.
> 
> So far, I have been trying to use the Gromacs user guide to extend the
> ethanol tutorial
> http://www.gromacs.org/@api/deki/files/196/=fe-tutorial-4.6.pdf
> into such a "dual" transformation, but also because I am currently faced
> with manually manipulating my topology file, this option is highly prone
> to my own errors.
> 
> Would you be able to help?
> 
> With Thanks and Best Regards,
> Sebastian
> --
> Gromacs Users mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.
> 


-- 
----------------------------------------------
Sebastian Stolzenberg, Ph.D.
PostDoc, Computational Molecular Biology group
Freie Universität Berlin

Phone: (+49) (0)30 838 75153
Mail:  Arnimallee 6, 14195 Berlin, Germany
----------------------------------------------