From: Vermaas, Joshua (Joshua.Vermaas_at_nrel.gov)
Date: Wed May 16 2018 - 17:45:13 CDT

Hi Fan,

You probably stumbled into psfgen finding a duplicate residue, and therefore they didn't get put back into the right place after psfgen solvated the system. What I would do is load up the unsolvated structure and the new solvated structure, and then copy the coordinates for everything that isn't water from the unsolvated structure to the solvated structure. So something like:

set unsolvated [atomselect 0 "all"]
set solvated [atomselect 1 "not water"]
foreach el [list x y z] {
$solvated set $el [$unsolvated get $el]
}

With any luck, there won't be any waters that you need to remove for being too close.

In terms of the ordering when topogromacs writes a file, I think this can be easily solved if you parenthesized your asel selection. "not type Kp or resname TIP3" groups like "(not type Kp) or (resname TIP3)", which is not the selection you want. Try "not (type Kp or resname TIP3)". Otherwise your modified script looks ok!

-Josh

On 2018-05-16 16:32:06-06:00 owner-vmd-l_at_ks.uiuc.edu wrote:

Hi

I did manage to add water molecules, but unphysical bonds are still there.

There is a atom at the origin, and lots of strange bonds are connected to it ,but it was there before I add water molecules because it is part of molecule2.

Moreover, I am using Josh's script to reorder my structure (with water) in the post http://www.ks.uiuc.edu/Research/vmd/mailing_list/vmd-l/29671.html08d5bb7cd9b1%7Ca0f29d7e28cd4f5484427885aee7c080%7C0%7C0%7C636621067265196542&sdata=osHeg9wWS%2ByxxhaNRdyWIoP5G35j19lFCnrcN8S8A1Y%3D&reserved=0>
and the convert to .top file. I found that the final top lists molecules in the following way see below

#############################
[ system ]
; Name
vmdmolecule3

[ molecules ]
; Compound #mols
molecule0 1
molecule1 1
molecule2 1380
molecule3 1
molecule4 10
molecule5 1
molecule6 2370
molecule7 1
molecule8 10
molecule9 1
molecule10 1191
molecule11 472
########################
It looks weird because molecules who have number larger than 1 are all water molecules,so the structure seems to be separated by water.

Then I modified the script a little bit
#########################################
topo cleardihedrals

#sort the structure and abandon the repeated fragments
set fragsellist [list ]
set asel [atomselect top "not type Kp or resname TIP3"]

foreach frag [lsort -unique [$asel get fragment]] {
set fsel [atomselect top "fragment $frag"]
lappend fragsellist $fsel
}

set midsel [atomselect top "resname TIP3"]
lappend fragsellist $midsel

#append K+ at the end of the pdb and psf files
set potsel [atomselect top "type Kp"]
$potsel set name K
lappend fragsellist $potsel
set newmol [::TopoTools::selections2mol $fragsellist]

pbc set [list $xmax $ymax $zmax]

animate write psf AFM_reordered.psf $newmol
animate write pdb AFM_reordered.pdb $newmol

topo writegmxtop AFM_reordered.top [list **.prm]
##########################################################

the results are more weird ,see below.
#############################################
[ system ]
; Name
vmdmolecule3

[ molecules ]
; Compound #mols
molecule0 1
molecule1 1
molecule2 1380
molecule3 1
molecule4 10
molecule5 1
molecule6 2370
molecule7 1
molecule8 10
molecule9 1
molecule10 6152
molecule11 472
#################################################

Can someone tell me what is wrong?

Thanks a lot.
Fan

2018-05-16 2:23 GMT+01:00 Dallas Warren <dallas.warren_at_monash.edu<mailto:dallas.warren_at_monash.edu>>:
Can't help with the first 1/ 2/, however with 3/ vmd generates bonds
to display based on the distance that the molecules are apart, not due
to any bonds present within a topology.
Catch ya,

Dr. Dallas Warren
Drug Delivery, Disposition and Dynamics
Monash Institute of Pharmaceutical Sciences, Monash University
381 Royal Parade, Parkville VIC 3052
dallas.warren_at_monash.edu<mailto:dallas.warren_at_monash.edu>
---------------------------------
When the only tool you own is a hammer, every problem begins to resemble a nail.

On 16 May 2018 at 07:59, fan li <fanliqmul_at_gmail.com<mailto:fanliqmul_at_gmail.com>> wrote:
> Hi every one
>
> I have molecule1 in the middle of my box, and molecule2 at the bottom. Now I
> want to fill the other volume not occupied by molecule1 and molecule2. I
> tried following command
>
> solvate ubq.psf ubq.pdb -t 5 -o ubq_wb
>
> It created the water around the molecule1 and molecule2, but the molecule2
> is already at the boundary of my box,so there should be no waters below it.
>
> (1)How can I fill the other volume not occupied by molecule1 and molecule2
> in vmd?
>
> (2)And what is the density of the solvate?
>
> (3)I also noticed that my original psf might be changed after I used the
>
> solvate ubq.psf ubq.pdb -t 5 -o ubq_wb because some non-physical bonds is
> generated in the display window? Why it happend?
>
> Fan