> I'd like to use AmberTools on my Desmond trajectories, so I need to
> convert the Desmond trajectory to Amber prmtop+mdcrd files.
Depends what you want to use in AmberTools as to whether you will need a
prmtop or not. For example, for many ptraj actions you only need a
corresponding PDB. However, if you are trying to do MM-PBSA then you need
a prmtop.
> To get mdcrd, I uploaded my Desmond trajectoy to VMD and saved all the
> frames as mdcrd by picking the file type as crdbox in VMD.
I would recommend saving them as a DCD (and then later converting to
crdbox if you need it post imaging, etc with ptraj). Smaller/faster/higher
precision.
> To get prmtop, I saved only one frame from the desmond trajectory as a
> pdb file in VMD. I was hoping to generate the prmtop using this pdb with
> tleap. However, as I load the pdb, tleap adds 2 terminal oxygens (OXT)
> to each of the chains in my protein-peptide complex.
If the protein and peptide are in separate molecules (i.e. not covalently
connected) they *will* have terminal OXT atoms unless appropriately
blocked (NME). If they do not in your Desmond runs, it suggests that
internal residues were used as terminal residues which is broken behavior,
i.e. you did not have COO- termini but CO-dangling bond with funny
non-integral charge.
> Is there a way to prevent tleap from adding these OXT atoms to the
> topology file?
Not so easily since we expect to have appropriate termini, but you
could try:
(1) removing the TER cards and let LEaP bond them together (this will only
remove one OXT).
(2) create a new terminal residue or remove the rules that build terminal
residues differently. CGLY = GLY, etc. In the leaprc file, there is a
section that specifies how to treat termini; edit this section
"addPdbResMap" noting that you will be breaking away from correct
behavior. i.e. for the C-terminal, that is the second section with 1 and
map GLN to GLN for example. Note only will this be broken, you will no
longer have an integral net-charge on the system.
> If not, I'm assuming I'll have to edit the topology file manually. In
> that case, would it work if I decreased the NATOM, MBONA, MTHETA, MPHIA
> by 1 in the POINTERS FLAG and deleted OXT from ATOM_NAME FLAG?
> (according to the topology file format described in
> http://ambermd.org/formats.html)
Do not even think about editing the prmtop by hand; much too
difficult as many things are coupled.
Note that running in AmberTools only makes sense for energetics if you
have a force field that matches exactly what you ran on/with Desmond.
--tec3
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Nov 02 2011 - 16:30:02 PDT
