Solute KE

From: Shantenu Jha (s.jha_at_ucl.ac.uk)
Date: Thu Dec 04 2003 - 16:36:31 CST

Hi,

I'm trying to measure the Temperature and Kinetic Energy of the solute
*only*, using NAMD 2.5. As I understand, in earlier versions of NAMD, it
wasn't possible; is it possible to do so with 2.5?

I tried to adapt the technique as given in the following link:

http://www.ks.uiuc.edu/Training/SumSchool/materials/sources/tutorials/02-namd-tutorial/namd-tutorial-html/node13.html#SECTION00031500000000000000

If as suggested, one generates a dcd file containing only solute (here
protein) atoms, wouldn't a modified coor and vel file be required
with the same number of atoms? (as well as modified input structure file
i.e. psf or its AMBER top equivalent, but this is easily handled).

Any suggestions?

Unless there is something I'm missing, can't the functionality of
the "time-consuming" mk_protein_dcd.tcl be easily emulated by
an appropriate indexfile entry using catdcd?

So although, I wasn't able to get the above approach working, using
another trick (involving generating a pdb file with the solvent atoms beta
value set to 1, and setting fixedAtoms parameter in my NAMD config file to
be "on" and setting fixedAtomsfile to the generatd pdb file), I was able
to get NAMD to spit out what appears to be solute only Kinetic Energies
(based on it passing some basic consistency checks), and thus back to the
first question, i.e. does NAMD 2.5 implement the ability to give KE/Temp
for a subset of atoms and thus can the numbers I get from the above
procedure be trusted?)

I guess I could rummage through the source code and answer parts of this
email, but I thought the question(s) might be of relevance to the wider
community.

Thanks.
Shantenu

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:37:12 CST