pKa calculation with thermodynamic integration (TI)

From: Sadegh Faramarzi Ganjabad (
Date: Mon May 08 2017 - 14:59:06 CDT

Dear all,

I am planning to calculate pKa of a Glu in a membrane protein with
thermodynamic integration method. As you know, CHARMM 36 has parameters for
both protonated and deprotonated Glu. However, there is no NAMD tutorial on
TI. I had few questions about my system setup. I assume that the dual
topology should be the same as free energy perturbation (FEP). And the
following terminal group of Glu needs to be changed during the perturbation

ATOM CG CT2 -0.21 !
ATOM HG1 HA2 0.09 ! HG1 OE1
ATOM HG2 HA2 0.09 ! | //
ATOM CD CD 0.75 ! -CG--CD
ATOM OE1 OB -0.55 ! | \
ATOM OE2 OH1 -0.61 ! HG2 OE2-HE2
ATOM HE2 H 0.44 !

Then, is this group supposed to vanish during the perturbation and a new
group should appear at the same position, with HE2 omitted and updated
partial charges for the rest of atoms? or only HE2 vanished and the charges
of other atoms are updated without vanishing/appearing? for the latter I am
not sure what the dual topology would look like (with keeping the atoms at
same positions and only reassigning partial charges).

Also, I am not sure what differences would be between FEP and TI in terms
of simulation procedure. I know the theory of each method, but compared to
FEP, there is no clear instructions about running TI as I said. In
'fep.tcl' script of FEP tutorial files there is a section for TI but I
don't know how to use it. Any help is highly appreciated.


This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:20:16 CST