From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Mon Aug 06 2012 - 10:57:36 CDT
I would like to explore the feasibility of your suggestion below in a
rather complex case of a whole protein that reacts with a drug,
incorporating a portion of that.
To this concern, may I ask what do you mean in practice with "write an
amber topology including both of my states"? Perhaps (a) writing
prmtop with LEaP for the two PDB pertaining to the two states of the
protein, or (b) writing prmtop for each state of the protein and then
appending a prmtop to the other one, or (c) what else?
At the end, one has to be careful that the number of atoms in the
final prmtop file - no matter how obtained - correspond to those in
the PDB file. At least, this holds for the AMBER code. Is that not so
for PARM7 as implemented in NAMD?
As to why not using psf/pdb directly, it is because I am better
equipped with PARM7 to deal with non-proteic stuff.
Thank you very much
FEP using amber topologies
Date: Mon Apr 23 2012 - 17:46:17 CDT
There is a very old post in the list asking this question just around
the time when the alchemify program was no longer needed to run FEP
simulations. So, as far as i understand if i write an amber topology
including both of my states in the same file and then i create the pdb
with the definitions of both end states then NAMD will be able to
infer the exclusion list even when you ask it to read the exclusion
list from the prmtop file (as required for the AMBER files).
Has anybody tried this? Or do any of you know if this approach works?
The other chance is to use one of this CHARMM formatted amber residue
topology files, but i'd rather use the files created by AmberTools
This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:21:54 CST