Re: questions regarding ABF sampling in a DMPC bilayer

From: Ajasja Ljubetič (
Date: Fri Jan 27 2012 - 08:03:54 CST

Dear Jrme and Giacomo,

> Good questions! I hope other users find them interesting.

Thanks for the answers! In a moment of weakness I was about ready to
throw away my PHD, and start bulidng
But things are making sense again:)

>> - Why are there such small energy differences in PMF between water an
>> DMPC?
>> To be more precise, things that appear as slow diffusion (i.e. friction)
> in the reaction coordinates are actually fluctuations in other degrees of
> freedom that couple to the RCs. Barriers in those directions, if they are
> too high, will kill diffusion in the RCs.

I always said: ok the transition rate over a barrier is proportional to
exp(Ea/(kT)), but forgot that diffusion can come into the equation as
a pre-exponential factor. So this is why I was surprised to find that the
rate of transition is different for the two systems, but the
energy barriers are very similar.

>> - How are gradients merged using
>> Are you sure the color scale is the same everywhere?
Indeed I was not. Mathematica was not doing exactly what I expected.

>> - Is it possible to use Accelerated MD with ABF (and get correct
>> results)?
>> It could be fun to try, but it might not be worth your time.
Sadly you are spot on: I would love to try, but with one year until my PHD
is due, time is extremely precious.

If you model the unbiased process as 2D diffusion on an effective potential
> (i.e. the PMF), then ABF will (in time) erase the barriers, but not change
> the intrinsic diffusion properties.

Actually what I would like to do now is to model the (unbiased) dynamic
behaviour of my colvars. I would like to know how far my colvars move in 10
ns. I plan to preform a random walk on the obtained PMF, where each step
will be accepted according to the Metropolis criterion (ie the probability
of accepting the step is min(1,exp(dE/(kT)))). Since such simulations are
computationaly extremely cheap I could get excellent sampling. The only
thing I have to figure out is what is the (real) time of one step is. *Am I
correct in guessing that it is proportional to
the autocorrelation time obtained from the colvars trajectory?*

Thank you and best regards,

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:24:41 CST