On Mon, Jul 16, 2012 at 9:57 PM, Osman, Roman <roman.osman.mssm.edu> wrote:
> I posted a question last week but no response, which leads me to conclude
> that perhaps I did not formulate it properly.
> Here is my problem, perhaps someone can help.
> I would like to calculate the energy of binding of a peptide to a protein
> using MMPBSA. I use two 100 ns trajectories of the protein-peptide complex
> and of the empty protein to create two new trajectories with 100 waters
> closest to the peptide binding groove of the peptide, respectively. I
> wrote them in ptraj with the nobox option. In the MMPBSA calculation I
> would like to use these trajectories but strip the waters from them. In
> addition, I have the trajectory of the peptide without water.
> I created the appropriate topologies of the solvated complex and empty
> protein (each with the waters) as well as the topologies of the complex,
> protein and peptide without waters. Now I run the MMPBSA with the three
> trajectories and solvated and non-solvated topologies. The run stops with
> the following error:
> .....
> TrajError: Solvated topology 0 has IFBOX == 0
> Error occured on rank 0.
> Exiting. All files have been retained.
> ....
>
For MMPBSA.py, solvated topology files are expected to be prepared with
tleap such that it is recognized as a solvated topology file with box
information. Therefore, MMPBSA.py will not generate stripped trajectory
files unless this is true. The main reason this is done this way is
because imaging is different for ifbox=1 (orthorhombic boxes) and ifbox=2
(triclinic boxes).
My suggestion is to generate the stripped trajectories yourself and use
that as the topology file. Then, do not supply a solvated topology file on
the command-line (this tells MMPBSA.py that you have already stripped the
trajectory and it is ready to use as-is).
Make sure you visualize the resulting trajectory with your complex topology
file to make sure everything looks OK. If so, this error should go away.
HTH,
Jason
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 16 2012 - 19:30:02 PDT
