Re: Unfavorable anti conformation after GLUP patching?

From: Sebastian Stolzenberg (s.stolzenberg_at_gmail.com)
Date: Thu Apr 16 2009 - 11:00:17 CDT

Hi Sébastien,

thank you, your definition of the syn/anti conformers made it all clear
to me now. All the GLUP and E2EP residues do start in the favorable syn
conformation. :-)

In response also to your other e-mail, you can either
(a) gradually increase/decrease the syn/anti barrier
(b) gradually increase/decrease a restraint on the dihedral
HE2-OE2-CD-OE1 to stay in the favorable syn conformation
I favor (b), see below.

Why do I write "increase/decrease"?

The total free energy difference will be
\Delta_G_tot=
\Delta_G(gradually increase)+
\Delta_G(FEP transformation, with restraint)+
\Delta_G(gradually decrease)

So far, it seems that you *abruptly* set the barrier to a higher value,
ignoring the \Delta_G(gradually increase/decrease) values, focusing on
\Delta_G(FEP transformation, with restraint) only. You might be lucky
though and still get the same \Delta_G_tot. But in general, it is
advisable to follow the scheme mentioned above.

How can one measure \Delta_G(gradually increase/decrease)?
Instead of aiming for option (a), I would do (b) with the PMF module of
NAMD >=2.6 and its "growth" command. Thus, you calculate \Delta_G by
slowly raising the force constant on the dihedral. Don't ask me what
"slow" means :-), usually people look at convergence behaviors. In
another case, I raised a positional restraint force constant on a single
atom to 10kcal/mol within 1ns for a total system with ~100K atoms (PMF
script attached).

Let me know, if that helps.
Bonne journee,
Sebastian


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:37 CST