[AMBER] pmemd.CUDA vs pmemd.MPI for equilibration of protein-ligand and trajectory identification for mmpbsa

From: Prasanth G, Research Scholar <prasanthghanta.sssihl.edu.in>
Date: Sat, 6 Apr 2019 12:17:01 +0530

Dear all,

I am interested in carrying out mmpbsa of a protein - ligand complex.
Question - 1:
I am interested in knowing if the equilibration (NVT and NPT) carried out
using pmemd.CUDA and pmemd.MPI would be one and the same.
If one of them is preferred, what is the reason?

With a bit of searching and going through the tutorials, I had carried out
the simulation in this process:
1) minimization -
a) water and counter ions (ncyc=1000, maxcyc=5000)
b) for all hydrogens (ncyc=1000, maxcyc=5000)
c) for alpha carbons (ncyc=1000, maxcyc=5000)
d) for all atoms (ncyc=1000, maxcyc=5000).

2) Heating through 0 K to 300 K with a rise of 50 K every 5000 steps for
30000 steps and mainatining the system at 300K for 10000 steps with
restraint weight of 5.0 on alpha carbon
3)NVT (nstlim = 20000, dt= 0.001) with no restraints
4)NPT (nstlim = 400000, dt= 0.001) with no restraints.
5) Production run (will use pmemd.CUDA) for 30 ns

Question - 2:
I would like to know, how to identify the region of the trajectory for
carrying out mmpbsa. I have read that we can do this by analyzing the
protein rmsd.

I would be grateful if you can point me to some literature concerning this

Thank you.

AMBER mailing list
Received on Sat Apr 06 2019 - 00:00:03 PDT
Custom Search