NAMDEnergy: run with XSC file or not?

From: Evgeny Bulat (jack.bulat_at_gmail.com)
Date: Wed Nov 21 2012 - 13:27:23 CST

Hey NAMD users,

I have a technical question regarding running NAMDEnergy to obtain
non-binding interactions between two molecules. I start VMD remotely via
command line and load up the appropriate psf and dcd files. After that, I
run a script that executes the following command for a bunch of pairs of
molecules in my system:

namdenergy -nonb -sel $sel1 $sel2 -ofile $output -tempname namd_nag1
-timemult 1 -stride 10000 -par [a bunch of param and toppar files] -pme
-extsys $xsc_file

NAMDEnergy outputs really nice and helpful energy graphs. However, when I
looked through some, I noticed that the initial binding energy is
occasionally really high, as in 60-70 kcals/mol! When I tried to figure out
why, one of the things I tried is running the above command without
specifying -extsys. That is:

namdenergy -nonb -sel $sel1 $sel2 -ofile $output -tempname namd_nag1
-timemult 1 -stride 10000 -par [a bunch of param and toppar files] -pme

Both are legit ways of running NAMDEnergy according to the manual. This
time, though, I got reasonable energies for the same pairs of molecules
that gave me really high, positive energies before. Just to clarify, the
xsc file I use is the one that preceeds the equilibration for which I'm
running NAMDEnergy.

Could anyone give me some guidance as to why specifying -extsys makes such
a big difference? Also, in light of this, which calculation should I trust
more?

Thank you!

Jack

-- 
Jack Bulat
Dept. of Molecular and Cell Biology,
Chemical Biology Graduate Program,
UC Berkeley

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:22:17 CST