I need to calculate the free binding energy between my protein and substrate but I need to know if this script works on dcd trajectories produced by NAMD or not.
Also, how can I decompose the calculated energy to residue contributions?
Thanks for any suggestions.
MMPBSA.py has no problem with NAMD .dcd files. It uses cpptraj to separate the trajectory into receptor and ligand components, and then outputs these trajectories as either .mdcrd or .nc files, depending on how you set the flag in your input script (I highly recommend using compressed .nc files to save space). Therefore, if cpptraj can parse the file type, then you can use it as an input to MMPBSA.py. However, it DOES require you to use stripped .prmtop files and not a CHARMM-formatted PSF. If you had used a PSF to conduct your simulations, you can build the system .prmtop information using tleap (or xleap), which is freely available as a part of AMBERTOOLS (just like MMPBSA.py). The online tutorial (http://ambermd.org/tutorials/advanced/tutorial3/py_script/) has the process for building the necessary stripped .prmtop files.
As for decomposing the output, that information is well documented in the AMBER manual. You include the "&decomp" section (with the desired formatting options) to your input script.
NAMD and the relevant post-simulation analysis software VMD are naturally compatible with Tk/Tcl scripts. One way to do this is to translate your .py scripts into Tcl scripts. But I do see someone running Python scripts for analyzing NAMD output files. namd-l mailing list would be another place where you can find the answer.
AMBER has detailed tutorial about MMPBSA.py. You could use CPPTRAJ to convert to MDCRD. But I have no clue how to deal with the PRMTOP file. Since MMPBSA.py reuqires proper radius properties for implicit solvent calculation. Just build your system with AMBER. You don't need a very long simulation for MMPBSA calculation.
Thank you for your answers.
@Xiaoquan Sun, I do not have access to Amber package.
MMPBSA.py has no problem with NAMD .dcd files. It uses cpptraj to separate the trajectory into receptor and ligand components, and then outputs these trajectories as either .mdcrd or .nc files, depending on how you set the flag in your input script (I highly recommend using compressed .nc files to save space). Therefore, if cpptraj can parse the file type, then you can use it as an input to MMPBSA.py. However, it DOES require you to use stripped .prmtop files and not a CHARMM-formatted PSF. If you had used a PSF to conduct your simulations, you can build the system .prmtop information using tleap (or xleap), which is freely available as a part of AMBERTOOLS (just like MMPBSA.py). The online tutorial (http://ambermd.org/tutorials/advanced/tutorial3/py_script/) has the process for building the necessary stripped .prmtop files.
As for decomposing the output, that information is well documented in the AMBER manual. You include the "&decomp" section (with the desired formatting options) to your input script.
I have recently used ambertools to perform mmpbsa for my system.
Steps followed:
1) Generate your PSF file using whatever tool necessary.
2) Run a simulation in NAMD with this PSF file.
3) Use chamber to generate a topology file compatible with cpptraj, sander,
and pmemd from the PSF file you generated in step 1. (You also need to
create the complex, receptor, and ligand topology files)
I have used chamber in parmed.py since there was a bug-gfortron error when I have used chamber to generate .prmtop and .inpcrd.
command:$amberhome/parmed-master/parmed.py
after running above command chamber is utilized as stdin as follows.
>chamber -top topology.top -prm parameter.prm -psf protein.psf -crd protein.pdb
>outpram output.prmtop output.inpcrd
>go
4) Run MMPBSA.py using the chamber-generated topology files you created in
step 3 with the DCD file you created in step 2.
command : mpirun -np 2 MMPBSA.py.MPI -O -i mmpbsa.in -cp com.top -rp rec.top
-lp lig.top -y traj.crd
Problem:
They do not allow us to calculate entropy using Nmode analysis,which is must for total free energy calc.
deltaG=deltaG-TdeltaS
PS - For detailed version go through ambertools14 manual(chamber,mmpbsa.py).
As Vijay mentioned, if you are not able to perform Nmode analysis, you may try doing Quasi Harmonic Approximation for entropy calculations. This need a large number of input frames. And the MMPBSA script take NAMD trajectories DCD as input trajectories.
AMBER14 is free now, unless you want to use GPU to speedup the simulation.
@Xiaoquan, AMBERTOOL14 is free to use not AMBER14 as it requires licence.
AMBERTOOLS14 contains SANDER to run MD simulation. You may comply it with MPI. I have made that on the HPCC and used it everyday. You may also use MMPBSA.MPI to calculate MMPBSA. I just finished that since entropy cost too much time. Try it before you make decision.
Please see my install note at https://sunxiaoquan.wordpress.com/2015/03/07/compile-ambertools14-on-ahpcc/
- Badges
- Science topic
- Similar topics
- Chemistry
- Theoretical Chemistry
- Computational Chemistry
- Molecular Dynamics