From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Tue Jun 15 2010 - 10:31:42 CDT

On Tue, Jun 15, 2010 at 11:14 AM, Rabab Toubar <rtoubar_at_yahoo.com> wrote:
> Using the tcl script you sent me, I am getting 11628 atoms, bonds present, & the protein is charged but not the water (I believe I'm getting the protein charges from the top file)
> 11628 !NATOM
>       1 MAIN 1    ASP  N    NL     0.129000       14.0067           0
>       2 MAIN 1    ASP  H1   H      0.248000        1.0080           0
>
> modifying the .tcl script so as to add ions that are found in my simulations, I am losing bonds and charges but having the correct number of atoms 11632..So I am just curious what is going on.
>
> #load .gro and .psf file into vmd
> set mol1 [mol new myfile.psf waitfor all]
> set mol2 [mol new myfile.gro waitfor all]
> set sel1 [atomselect $mol1 all]
> set sel2 [atomselect $mol2 water]
> set sel3 [atomselect $mol3 ion]

this one cannot work. there is no $mol3.

set sel3 [atomselect $mol2 ion]

to reset the charges and/or other properties,
all you have to do is to build the proper selections
and then set the values before combining them.
for example.

set sel4 [atomselect $mol2 "type OW"]
$sel4 set charge -0.84
set sel5 [atomselect $mol2 "type HW"]
$sel5 set charge 0.42

and so on.

cheers,
   axel.

> # merge selections into new molecule.
> package require topotools
> ::TopoTools::selections2mol "$sel1 $sel2 $sel3"
>
> # write out new molecule
> animate write psf combined.psf
>
> 11632 !NATOM
>       1      1    ASP  N    N      0.000000       14.0070           0
>       2      1    ASP  H1   H1     0.000000        1.0080           0
>
> Thanks, bondsrecalc made the bonds appear in the psf, and sorry for the email format & the cc; it was unintended.
>
> Rabab
>
> --- On Tue, 6/15/10, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>
> From: Axel Kohlmeyer <akohlmey_at_gmail.com>
> Subject: Re: vmd-l: top2psf.tcl
> To: "Rabab Toubar" <rtoubar_at_yahoo.com>
> Date: Tuesday, June 15, 2010, 10:48 AM
>
> On Tue, Jun 15, 2010 at 10:38 AM, Rabab Toubar <rtoubar_at_yahoo.com> wrote:
>>
>> Hi Axel,
>>
>> using "animate" I get the correct number of atoms, but all the charges are set to zeros, and I have no bonds
>> 11632 !NATOM
>>        1      1    ASP  N    N      0.000000       14.0070           0
>>        2      1    ASP  H1   H1     0.000000        1.0080           0
>>        3      1    ASP  H2   H2     0.000000        1.0080           0
>
> well, you cannot have what is not there. a .gro file has
> not information about atom charges, so there are not
> in the resulting .psf file.
>
>> 0 !NBOND: bonds
>>        0 !NTHETA: angles
>>       0 !NPHI: dihedrals
>>        0 !NIMPHI: impropers
>>       0 !NDON: donors
>>       0 !NACC: acceptors
>>
>> the second "untested" method worked well but not including the ions. It had the charges for the atoms and everything. So I tried to modify the tcl script by adding a 3rd selection "set sel3 [atomselect $mol3 ion]" when I got zero charges and zero bonds as the 1st case. I donot understand why; any guesses?
>
> it it not clear what you did.
>
>
> not sure why you are not getting any bonds.
> you can try loading the .gro file and then do
> mol bondsrecalc $mol
> topo retypebonds
>
> that should trigger the internal flags in VMD
> to write out bonds.
>
> you have to manually set any charges that are not
> given through input in case you want them set in the .psf.
>
> you have to remember this is not psfgen where this is
> all read from a parameter database, you are doing all
> steps by hand.
>
> cheers,
>    axel.
>
> p.s.: please turn off html format mails to mailing lists and always
> keep the list in cc.
> i usually don't respond to questions send to me directly that don't concern me
> personally, or else i would have no time to get anything done. thanks.
>
>> However, when I added the Cl- manually the psf generation went well and loaded in VMD
>>
>> Thanks a lot for your help :)
>>
>> Rabab Toubar
>>
>>
>> --- On Tue, 6/15/10, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>>
>> From: Axel Kohlmeyer <akohlmey_at_gmail.com>
>> Subject: Re: vmd-l: top2psf.tcl
>> To: "Rabab Toubar" <rtoubar_at_yahoo.com>
>> Cc: "vmd-l" <vmd-l_at_ks.uiuc.edu>
>> Date: Tuesday, June 15, 2010, 9:59 AM
>>
>> On Tue, Jun 15, 2010 at 9:49 AM, Rabab Toubar <rtoubar_at_yahoo.com> wrote:
>> >
>> > I need the .psf file as we have in house codes used for analysis and these use .psf and .dcd as in case of NAMD.
>>
>> > I tried to load the .gro or .pdb files and save coordinates as psf, but this is not an option in the drop down menu in vmd; is there a way I can do it in tcl?
>>
>> of course:
>>
>> animate write psf myfile.psf
>>
>> > I do have disulfide bonds- if this what you meant by unusual bonds.
>> >
>> > In the meantime I will try in gromacs to get the .top file from the .gro and see if this would work to get the .psf.
>>
>> one possible alternative (untested):
>>
>> #load .gro and .psf file into vmd
>> set mol1 [mol new myfile.psf waitfor all]
>> set mol2 [mol new myfile.gro waitfor all]
>> set sel1 [atomselect $mol1 all]
>> set sel2 [atomselect $mol2 water]
>>
>> # merge selections into new molecule.
>> package require topotools
>> ::TopoTools::selections2mol "$sel1 $sel2"
>>
>> # write out new molecule
>> animate write psf combined.psf
>>
>>
>>
>>
>>
>>
>> > Thanks
>> > Rabab Toubar
>> >
>> >
>> > --- On Tue, 6/15/10, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>> >
>> > From: Axel Kohlmeyer <akohlmey_at_gmail.com>
>> > Subject: Re: vmd-l: top2psf.tcl
>> > To: "Rabab Toubar" <rtoubar_at_yahoo.com>
>> > Cc: "vmd-l" <vmd-l_at_ks.uiuc.edu>
>> > Date: Tuesday, June 15, 2010, 9:29 AM
>> >
>> > On Tue, Jun 15, 2010 at 9:21 AM, Rabab Toubar <rtoubar_at_yahoo.com> wrote:
>> > >
>> > > Thanks Axel for your reply
>> > >
>> > > My .top file has no waters, as it is the initial .top file I get using grompp, and it has a total of 357 atoms (attached).
>> > >
>> > > The md gro file has 11632, and it contains waters.
>> >
>> > well, you cannot load multiple files into the same VMD
>> > "molecule" that have different number of atoms.
>> >
>> > > I can visualize the molecule using .pdb or .gro file with the .trr, but I need to have a psf files for further analyses, so can I possibly include waters in the top file? or may be delete it from the .trr file?
>> >
>> > what exactly do you need the .psf file for?
>> >
>> > the one generated from the .top file is only minimal
>> > and only contains the bonding information and atom
>> > names. that is suitable for visualization in VMD, but
>> > not much more. in most cases, you can generate the
>> > same kind of file by loading a .gro or .pdb file and saving
>> > the result in a .psf file. only if you have unusual bonding,
>> > you'd need to read it from the .top file.
>> >
>> > cheers,
>> >     axel.
>> >
>> >
>> >
>> > >
>> > > Thanks
>> > > Rabab Toubar
>> > >
>> > >
>> > > --- On Sun, 6/13/10, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>> > >
>> > > From: Axel Kohlmeyer <akohlmey_at_gmail.com>
>> > > Subject: Re: vmd-l: top2psf.tcl
>> > > To: "Rabab Toubar" <rtoubar_at_yahoo.com>
>> > > Cc: "vmd-l" <vmd-l_at_ks.uiuc.edu>
>> > > Date: Sunday, June 13, 2010, 8:19 AM
>> > >
>> > > > However when I load the psf file in vmd, I got the following:
>> > > >
>> > > > "psfplugin) WARNING: no angles defined in PSF file.
>> > > > psfplugin) WARNING: no dihedrals defined in PSF file.
>> > > > psfplugin) WARNING: no impropers defined in PSF file.
>> > > > psfplugin) no cross-terms defined in PSF file.
>> > > > Info) Analyzing structure ...
>> > > > Info)    Atoms: 357
>> > > > Info)    Bonds: 365
>> > > > Info)    Angles: 0  Dihedrals: 0  Impropers: 0  Cross-terms: 0
>> > > > Info)    Bondtypes: 0  Angletypes: 0  Dihedraltypes: 0  Impropertypes: 0
>> > > > Info)    Residues: 36
>> > > > Info)    Waters: 0
>> > > > Info)    Segments: 1
>> > > > Info)    Fragments: 1   Protein: 1   Nucleic: 0"
>> > >
>> > > that is to be expected after the debug output from above.
>> > >
>> > > > And when I load the trr file from GMX, I get the following:
>> > > >
>> > > > "ERROR) BaseMolecule: attempt to init atoms while structure building in progress!
>> > > > ERROR) Invalid number of atoms in file: 11632
>> > > > Info) Using plugin trr for coordinates from file /home/Desktop/gmx/pr_md_1.trr
>> > > > ERROR) Incorrect number of atoms (11632) in
>> > > > ERROR) coordinate file /home/Desktop/gmx/pr_md_1.trr
>> > > > Info) Finished with coordinate file /home/Desktop/gmx/pr_md_1.trr."
>> > >
>> > > either the .trr file doesnt match the .top file, or there is something
>> > > else wrong, that is impossible to debug without having access
>> > > to those files. have you tried loading a matching.gro file?
>> > > how many atoms does it have?
>> > >
>> > > cheers,
>> > >       axel.
>> > >
>> > > > PS I am using the preprocessed top file like Axel suggested in an earlier post
>> > > >
>> > > > Any suggestions would be much appreciated
>> > > > Thanks
>> > > > Rabab Toubar
>> > >
>> > >
>> > >
>> > > --
>> > > Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
>> > > http://sites.google.com/site/akohlmey/
>> > >
>> > > Institute for Computational Molecular Science
>> > > Temple University, Philadelphia PA, USA.
>> > >
>> > >
>> >
>> >
>> >
>> > --
>> > Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
>> > http://sites.google.com/site/akohlmey/
>> >
>> > Institute for Computational Molecular Science
>> > Temple University, Philadelphia PA, USA.
>> >
>>
>>
>>
>> --
>> Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
>> http://sites.google.com/site/akohlmey/
>>
>> Institute for Computational Molecular Science
>> Temple University, Philadelphia PA, USA.
>>
>
>
>
> --
> Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
> http://sites.google.com/site/akohlmey/
>
> Institute for Computational Molecular Science
> Temple University, Philadelphia PA, USA.
>
>
>
>
>
>

-- 
Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
http://sites.google.com/site/akohlmey/
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.