Re: Simulation of the membrane protein

From: Niklaus Johner (niki.johner_at_gmail.com)
Date: Thu Sep 05 2013 - 12:23:00 CDT

Simply using a tcl script and vmd. I think something like this works:

#read in the trajectory
mol new ionized.psf
mol addfile ionized.pdb

#make a selection and write out the psf and pdb
set sel [atomselect 0 "segid PROT"]
$sel writepdb prot.pdb
$sel writepsf prot.psf

N.

Niklaus Johner
Weill Cornell Medical College
Harel Weinstein Lab
Department of Physiology and Biophysics
1300 York Avenue, Room D-501
New York, NY 10065

On Sep 5, 2013, at 11:52 AM, James Starlight wrote:

> One extra question about solvation
>
> I've forced with the problem of the expanding of my system in Z dimension ( adding some addition solvent)
>
> In fact I've used membrane made by VMD and add some water by means of solvate plugin. Than I equilibration that hydrated bilayer and inserted protein into it. After visualization I've decided to add some more water to increase Z dimension of my system. Unfortunately second solvation of that membrane by means of solvate plugin ended with the error because of the overlapping segment names (WT1- WT10 produced during first solvation). How I could remove all old water (WT1-WT10) from both pdb and psf files and make new PSF for the de-solvated system ?
>
> James
>
>
> 2013/9/5 James Starlight <jmsstarlight_at_gmail.com>
> I've noticed the possible source of such error could arise from unproper minimization cycle.
>
> I've noticed that coordinates of the lipid and water didn't change during minimization cycle (although I didn't use any fix or restraints options in the conf file) such problem occurred only in case of the protein+solvent complex (in case of the minimization of the pure solvent the coordinates of the lipid tales as well as water change rapidly during minimization ).
>
> By the way does someone know possible source of the pre-equilibrated bi-layers made in charm params?
>
> Thanks for help
>
> James
>
>
> 2013/9/5 James Starlight <jmsstarlight_at_gmail.com>
> Aron, thanks for suggestions!
>
> I've done all such operations but the results the same.
>
> If it possible I can share pdb as well as psf files of my complex produced after ionization by VMD.
>
> Perhaps some problems might be also due to the insertion (automate generation of the complex psf which have been done according to the script provided in the http://users.mccammon.ucsd.edu/~rlaw/ctbp_workshop_rlaw.htm ) If someone have modified version of such script for quick insertion and deletion of the lipids in the aligned bi-layer I'll be thankful for it as well as for example conf file for production run of the protein-lipid system made with ch27 params
>
> Thanks again,
>
> James
>
>
> 2013/9/4 Aron Broom <broomsday_at_gmail.com>
> just on the PBC thing, although I somewhat doubt it's the answer. Your cellBasisVector values don't agree with what you posted from VMD. For instance, you have cellBasisVector3 as 98.2, but that value from VMD should be (93.664 - 0.594) which is 93.07, the others are off by much less.
>
> Some other things to try before the restraints:
>
> - increase minimization steps (go 10x what it currently is maybe)
> - decrease timestep (go to 0.5 for instance)
> - decrease temperature (start at something quite low like < 100K)
>
> Also, you said analysis doesn't reveal artifacts, but if there is a constraint failure, then something went flying off at high speed. So you might want to decrease you DCDFrequency or whatever the variable is called so you can see what is leading to the problem, particularly if this is happening early on.
>
> ~Aron
>
>
> On Wed, Sep 4, 2013 at 12:16 PM, James Starlight <jmsstarlight_at_gmail.com> wrote:
> Dear NAMD Users!
>
>
> Recently I've forced with the problem of simulation membrane protein system.
>
> I've done all steps according to the official tutorial
>
> 1) building POPE membrane using VMD
>
> 2) generated PSF files of my protein
>
> 2) inserted protein into membrane and removed overlapped lipids
>
> 3)solvated and ionized my system using VMD
>
> Than I've minimized my system and tried pack lipids but simulation have been crashed
>
> on this step I've froze all water ion and protein atoms and run short simulation in NPT according to the tutorial parameters.
>
> as the result I've obtained
>
> Info: ABSOLUTE IMPRECISION IN FAST TABLE ENERGY: 1.69407e-21 AT 11.9974
> Info: RELATIVE IMPRECISION IN FAST TABLE ENERGY: 1.13046e-16 AT 11.9974
> Info: INCONSISTENCY IN FAST TABLE ENERGY VS FORCE: 0.000290274 AT 0.251946
> Info: INCONSISTENCY IN VDWA TABLE ENERGY VS FORCE: 0.0040507 AT 0.251946
> Info: INCONSISTENCY IN VDWB TABLE ENERGY VS FORCE: 0.00150189 AT 0.251946
> Pe 2 hosts 15 local and 15 remote patches for pe 2
> Pe 3 hosts 15 local and 15 remote patches for pe 2
> Pe 1 hosts 10 local and 10 remote patches for pe 2
> Pe 0 hosts 23 local and 22 remote patches for pe 2
> Info: useSync: 1 useProxySync: 0
> Info: Startup phase 10 took 0.037595 s, 102.809 MB of memory in use
> Info: Startup phase 11 took 0.000133991 s, 102.973 MB of memory in use
> Info: Startup phase 12 took 0.000496864 s, 103.18 MB of memory in use
> Info: Finished startup at 1.20948 s, 103.34 MB of memory in use
>
> TCL: Running for 250000 steps
> Pe 2 has 63 local and 62 remote patches and 1701 local and 1674 remote computes.
> ERROR: Constraint failure in RATTLE algorithm for atom 35875!
> ERROR: Constraint failure; simulation has become unstable.
> ERROR: Exiting prematurely; see error messages above.
> ====================================================
>
> its remarkable that such error have occurred even with applied position restraints to ALL atoms of my system. In addition I've tried to make such packing for pure hydrated lipids but forced with the same error too.
>
>
> Might it be due to some problems with the PBC definition (I've done manually definition of the xyz dimensions using vmd as you can see below
>
> atomselect0
> >Main< (GPRC_namd) 2 % measure minmax $everyone
> {1.0085545778274536 0.6706096529960632 0.593999981880188} {104.46700286865234 102.0479965209961 93.66400146484375}
>
> measure center $everyone
> 52.83515548706055 51.502288818359375 46.488792419433594
>
> and than defined that in the conf file:
>
>
> cellBasisVector1 103.1 0 0
> cellBasisVector2 0 100.9 0
> cellBasisVector3 0 0 98.2
> cellOrigin 52.80777359008789 51.45378112792969 46.48970031738281
>
>
> during analysis of the minimization I've not observed any artifacts.
>
> What also membrane simulation options should I take into account?
>
>
> Thanks for help,
>
> James
>
>
>
>
>
>
>
> --
> Aron Broom M.Sc
> PhD Student
> Department of Chemistry
> University of Waterloo
>
>
>

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:21:37 CST