Free Energy Purturbation and Hamiltonian Replica Exchange

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Free Energy Purturbation and Hamiltonian Replica Exchange

Mfennim1
Greetings again,

Is there a way to introduce a coupling parameter to a solute in order to run Free Energy Perturbation or Hamiltonian Replica Exchange for solvation free energy or binding free energy calculations? I don't believe Abalone is currently set up for this, but I was wondering if there would be some way to run the calculations anyway?

If not, can you recommend some ofter method for determining binding and solvation free energies?

Thanks for the continual support. This is a wonderful simulations package.

Mark
Reply | Threaded
Open this post in threaded view
|

Re: Free Energy Purturbation and Hamiltonian Replica Exchange

agilemolecule
Administrator
Not yet, but this is in the plans.

However MDynaMix is able to calculate the free energy, and does it very efficiently.









Reply | Threaded
Open this post in threaded view
|

Re: Free Energy Purturbation and Hamiltonian Replica Exchange

mfennim
Thanks for the tip. I will certainly give this a try. In the mean time, I'm trying to perform calculations in the NPT (isothermal/Isobaric) ensemble. I'm currently thermally relaxing my system using the Berendsen thermostat, and performing my production runs using:

Berendsen Borostat (pressure set to achieve desired average density)
Nose-Hoover Thermostat
Velocity Verlet Integration @ 1fs MD step

Are these the correct parameters for the NPT?

Mark
Reply | Threaded
Open this post in threaded view
|

Re: Free Energy Purturbation and Hamiltonian Replica Exchange

agilemolecule
Administrator
mfennim wrote
Berendsen Borostat (pressure set to achieve desired average density)
Nose-Hoover Thermostat
Velocity Verlet Integration @ 1fs MD step

Are these the correct parameters for the NPT?
To be precise, Berendsen Barostat does not give NTP ensemble. However, for sufficiently large relaxation time difference is small. It seems to me that the Abalone default values are reasonable.

1 fs for Velocity Verlet is a limit.

In general, with such parameters we can work. But of course it is necessary to do validation.



Reply | Threaded
Open this post in threaded view
|

Re: Free Energy Purturbation and Hamiltonian Replica Exchange

mfennim
In reply to this post by agilemolecule
Thanks for the Tip on MDynaMix. I've seen good convergence between calculation and experiment even with fairly short simulation times (e.g. something like 3ns ).
kudos
Mark