Respected sir,
I would like to explain the whole scenario.
1) I had two PDB files of same protein with different terminal residues let us give them name PDB1 and PDB2
2) PDB1 has about 30 residues less than PDB2
3) I ran MD simulation for both using same protocol
4) I stripped extra residues thus having similar number of residues in both.
5) However, as the PDB2 had extra residues the water molecules were different in the trajectory and prmtop files after stripping which can be seen from the a portion of .out files
.....................................................
ACTION SETUP FOR PARM [complex_solvated.prmtop] (1 actions):
0: [strip :1,151-154,Na+,Cl- outprefix stripped_PDB1]
Stripping 74 atoms.
Topology complex_solvated.prmtop contains 23467 atoms.
7149 residues.
23490 bonds.
7001 molecules.
Box: Trunc. Oct.
6999 solvent molecules.
Final solute residue is 150
Writing out amber topology file complex_solvated.prmtop to stripped_PDB1.complex_solvated.prmtop
.....................................................
ACTION SETUP FOR PARM [complex_solvated.prmtop] (1 actions):
0: [strip :1-18,168-182,Na+,Cl- outprefix stripped_PDB2]
Stripping 476 atoms.
Topology complex_solvated.prmtop contains 42599 atoms.
13526 residues.
42622 bonds.
13378 molecules.
Box: Trunc. Oct.
13376 solvent molecules.
Final solute residue is 150
Writing out amber topology file complex_solvated.prmtop to stripped_PDB2.complex_solvated.prmtop
6) Now, I want to merge the stripped trajectories and do analysis on final trajectory like clustering of ligand poses, binding free energy etc.
7) However, for that I need a prmtop file and which is the place I am stuck.
8) I thought that by stripping extra water from the PDB2 prmtop and trajectory I can do the analysis.
I hope you can get an idea what I am trying to achieve and guide me accordingly. Please do let me know if I need to change the approach as the main aim is to correlate the pose and binding free energy of the ligand with IC50 value. The other idea that comes to my mind is to use both trajectories separately and see if the results are same in both in terms of pose and binding free energy trend.
Thanks and regards,
Nitin Sharma
-----Original Message-----
From: Jason Swails [mailto:jason.swails.gmail.com]
Sent: Wednesday, April 02, 2014 9:52 PM
To: AMBER Mailing List
Subject: Re: [AMBER] truncatiing water molecules
On Wed, Apr 2, 2014 at 9:03 AM, Nitin Sharma <sharmanitin.nus.edu.sg> wrote:
> Dear Jason,
>
> Following your instructions I successfully managed to strip extra
> solvent molecules.
>
> However, I am having a doubt and hope you can clarify it . As far as I
> know in PBC the molecules lost from one side are gained from other end.
> Hence, the water molecules lost from one end due to movement while
> simulating will appear from other end. This should result in
> shuffling of water molecules surrounding the protein. Now, as I have
> stripped water molecules i.e. from 7150-13526 is there a possibility
> that some of them were taking part in interaction in one or more trajectories?
>
The "strip" command will delete the atoms and all of their periodic images from the system. From the look of your command, you are deleting about half of your water molecules (!!) which will cause huge pockets of vacuum in your system.
I'm not sure what you mean by "is there a possibility that some of them were taking part in interaction in one or more trajectories?" -- every particle interacts with every other particle in molecular simulations. By deleting particles you are certainly getting rid of particles that _were_ interacting with other particles during the trajectory and creating unphysical vacuum pockets in the system. I'm still puzzled about why you're trying to match up the numbers of waters between two systems. If you need to carry out any energy calculations, though, the system from which you deleted ~6400 water molecules will be basically useless.
Good luck,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 02 2014 - 08:00:04 PDT
