Jarzynski-equation based steered MD

778 views
Skip to first unread message

tusar ban

unread,
Sep 29, 2010, 3:09:12 AM9/29/10
to plumed...@googlegroups.com
Dear Plumed-users,

I am new to PLUMED and have successfully installed GROMACS patched with PLUMED. In this regard, the manual and the  paper (CPC 180, 1961) was of great help.
I am interested in Jarzynski-equation based steered MD simulation of protein-drug interaction. Following the examples that the manual have put forward, I can set up the steered MD through the dat file :

# a simple CV: a dihedral angle
ANGLE LIST 13 15 17
# switching on steered MD
STEER CV 1 TO 3.0 VEL 0.5 KAPPA 500.0
# instruction for CVs printout
PRINT W_STRIDE 100
# end of the input
ENDMETA

In view of this dat file, I have two queries.
1. In addition to above, while executing Gromacs for steered MD do I also need to use the *.ppa file that one normally uses for Gromcas-only steered MD.

2. After the Gromcas-Plumed run for steered MD run finishes, how do I get the free energy profile? Is it through the sum_hills utility of Plumed? If so, then do I need to add the line:

HILLS HEIGHT 0.1 W-STRIDE 500

to the above general input dat file for Plumed steered MD?

I shall look forward for the answers. Thanks.
tusar
India

tusar ban

unread,
Sep 29, 2010, 3:16:44 AM9/29/10
to PLUMED users

davide branduardi

unread,
Sep 29, 2010, 8:05:02 AM9/29/10
to plumed...@googlegroups.com
Hi Tusar

1) the steer input is fine. That does not provide free energy but you could retrieve work values
  (rather different from free energies. See Jarzynski paper
    C. Jarzynski, Nonequilibrium equality for free energy differences, Phys. Rev. Lett. 78, 2690 (1997) )
The way to compute work:
   JOURNAL OF CHEMICAL PHYSICS    VOLUME 119, NUMBER 6    8 AUGUST 2003
Free energy calculation from steered molecular dynamics simulations using Jarzynski’s equality
Sanghyun Park et al
2) a HILLS directive won't work. This is useful only if you use metadynamics as free energy method.
   See the relevant literature from Laio &Parrinello and/or the CPC Plumed manifesto
3) A possible syntax for the steered MD is included in the tests along with plumed. check this dir in your
   plumed installation
   tests/gromacs/steer  
   there also you may find the files that are generally needed: a *tpr (standard for gromacs) and a plumed input (peculiar to
   plumed). Disable all the possible Steered MD features native in gromacs if you want to use plumed.

Hope this may help!
Ciao
D

 



--
You received this message because you are subscribed to the Google Groups "PLUMED users" group.
To post to this group, send email to plumed...@googlegroups.com.
To unsubscribe from this group, send email to plumed-users...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/plumed-users?hl=en.



--
**************************************************************
Davide Branduardi
Post Doc- Computational Chemistry
Drug Discovery and Development
ISTITUTO ITALIANO DI TECNOLOGIA
Via Morego 30, 16163 - Genova, Italy
Phone: +39 010 71781573              Room: 1/27
Fax:   +39 010 7170187

e-mail:
davide.b...@iit.it
davide.b...@gmail.com

tusar ban

unread,
Sep 30, 2010, 4:31:10 AM9/30/10
to plumed...@googlegroups.com
Dear Davide Branduardi,
Thanks for your advice. I will read the papers that you have suggested.
Regards.
tusar
Reply all
Reply to author
Forward
0 new messages