VMD-L Mailing List

From: Daniel Guion (danielguion_at_g.ucla.edu)
Date: Mon Dec 16 2024 - 16:35:57 CST

Hi Josh,

Thank you for the insights. However, I am confused by your message.

The residues in my pdb and psf files are not explicitly named ASPP, GLUP,
or LSN; they are named ASP, GLU, and LYS as per the standard convention.
The alternative protonation states (ASPP, GLUP, and LSN) were assigned by
PropKa, and their definitions already exist in the default RTF file used by
mutator. My understanding was that if the residue names remain standard
(ASP, GLU, LYS) and the correct connectivity exists for their modified
states, the patches in the RTF should handle these variations seamlessly.

Additionally, the mutator documentation suggests it supports user-specified
patches, as it automatically applies NTER and CTER patches when required. I
also noticed that the mutator TCL script parses input psf files for
user-specified patches and even includes comments hinting at plans for
custom topology files. Since my use case only involves the patches already
present in the default RTF, I was under the impression that adding the
relevant REMARKS patch line at the top of my psf file (with properly
matching segment IDs and residue numbers) should work without additional
intervention.

One option I can think of is taking the output psf and pdb from mutator and
adding back those specific protonation states from propka using charmm. Do
you think this is the best option? The reason I am not just working in
charmm to mutate my residue is that I need the hybrid topology for FEP
that the VMD mutator uniquely generates.

Best,
Daniel

On Mon, Dec 16, 2024 at 6:50 AM Vermaas, Josh <vermaasj_at_msu.edu> wrote:

> Hi Daniel,
>
>
>
> The ASPP, GLUP, and LSN patches are applied **on top** of existing ASP,
> GLU, and LYS residues. So giving residue names ASPP, GLUP, and LSN can’t
> work, since most of the atoms (like the whole protein backbone!) are not
> defined by the patch residue. The usual workflow is to make a
> “normal”/final psf/pdb pair, throw that into mutator to setup the mutation,
> and if needed fix up whatever is wrong afterward with additional patch
> commands. Mutator is not doing anything magical, and is just a tcl script
> that calls psfgen. It does have some side effects, since it is building a
> new segment, and that is what is blowing away the protonation state you’ve
> already assigned. If you want to keep using the plugin, you’d need to
> reapply the patches in psfgen after the fact from the VMD output.
>
>
>
> -Josh
>
>
>
> *From: *<owner-vmd-l_at_ks.uiuc.edu> on behalf of Daniel Guion <
> danielguion_at_g.ucla.edu>
> *Date: *Friday, December 13, 2024 at 6:31 PM
> *To: *"vmd-l_at_ks.uiuc.edu" <vmd-l_at_ks.uiuc.edu>
> *Subject: *vmd-l: Issue with patching when running mutator in vmd
>
>
>
> Hello,
>
> I am trying to run mutator on a solvated system with residues that have
> alternative protonation (ASPP, GLUP, and LSN; from propka), which already
> have patches in "/Applications/VMD
> 1.9.4a57-arm64-Rev12.app/Contents/vmd/plugins/noarch/tcl/readcharmmtop1.2/top_all36_prot.rtf
> <https://urldefense.com/v3/__http:/1.9.4a57-arm64-Rev12.app/Contents/vmd/plugins/noarch/tcl/readcharmmtop1.2/top_all36_prot.rtf__;!!DZ3fjg!7IC-WUIPFbB3rrah_B3X3u67FVw0qQsXTMBfYV-infbpzGh3177PcSMxQHzCWUrBwCEoq_9w1BhcWttGVDCB80rKkw$>".
> It makes sense to me that I don't have to create custom new patches in this
> file to specify them in the psf since they are relatively common
> alternative protonation states of standard protein residues.
>
>
>
> Here is how I specified an example patch at the top of my psf file
> (reduced to 1 patch for troubleshooting)
> " 2 !NTITLE
>
> REMARKS patch ASPP A3:137"
>
> When I run mutator, though, it crashes and I get this error
>
> " Mutator: Writing temporary files for non-protein components
>
> psfgen) Info: writing psf file mtemp-nprot.psf
>
> psfgen) total of 1054889 atoms
>
> psfgen) total of 1052154 bonds
>
> psfgen) total of 664357 angles
>
> psfgen) total of 526655 dihedrals
>
> psfgen) total of 3176 impropers
>
> psfgen) total of 0 explicit exclusions
>
> psfgen) Structure requires EXTended PSF format
>
> psfgen) total of 0 cross-terms
>
> psfgen) no residue 137 of segment A3
>
> /Applications/VMD 1.9.4a57-arm64-Rev12.app/Contents/MacOS/startup.command
> <https://urldefense.com/v3/__http:/1.9.4a57-arm64-Rev12.app/Contents/MacOS/startup.command__;!!DZ3fjg!7IC-WUIPFbB3rrah_B3X3u67FVw0qQsXTMBfYV-infbpzGh3177PcSMxQHzCWUrBwCEoq_9w1BhcWttGVDDkbRyN7w$>:
> line 7: 83723 Trace/BPT trap: 5 "$p/../Resources/VMD.app/Contents/MacOS/VMD"
> $*
>
>
>
> Saving session...
>
> ...copying shared history...
>
> ...saving history...truncating history files...
>
> ...completed.
>
>
>
> [Process completed]"
>
> If I remove the patch line from my psf file, VMD does not crash and
> instead runs to completion - however the script notes a warning that the
> coordinates for the extra hydrogen atom for that residue cannot be guessed.
> "psfgen) reading coordinates from pdb file mtemp-A3.pdb for segment A3
>
> psfgen) Warning: failed to set coordinate for atom HD2 ASP:137 A3
> ...
>
> Mutator: System net charge after mutation: -40.7995137386024e
>
>
>
> Mutator 1.3 completed successfully.
>
>
>
> WARNING: mutations were performed using psfgen. Original backbone
>
> atom coordinates were used for positions of backbone atoms of the
>
> new amino acid(s). Original coordinates of sidechain atoms named
>
> identically to atoms of the new amino acid(s) were used for positions
>
> of the new sidechain atoms. The positions of the remaining atoms
>
> were guessed using internal coordinates for each amino acid. It is
>
> therefore *highly recommended* to visually inspect the resulting
>
> structure and perform a minimization/equilibration of the system."
>
> When I look at the output psf and pdb file without adding the patch to the
> psf, it looks okay except I don't have the proton like in the input anymore
> - it removed it. Plus the charge of the system is off because I don't have
> the protonation I solvated (water+ions) with (-40.7995137386024e); recall I
> am just showing 1 alternative protonation state in this message for
> troubleshooting, but my whole system contains other alternative
> protonations (from propka).
>
> I then tried purposefully changing the patch line to have the wrong
> formatting by removing the ":"
>
> " REMARKS patch ASPP A3 137"
> instead of
> " REMARKS patch ASPP A3:137"
>
> And the mutator no longer recognizes that I have a patch specified in the
> psf, which was a nice validation that I have the correct patch notation, I
> suppose.
>
> I’ve also ensured that the residue numbers and segment IDs match between
> the PSF and PDB files
>
>
>
> PSF -
> 6958 A3 137 ASP N NH1 -0.470000 14.0070 0
> 6959 A3 137 ASP HN H 0.310000 1.00800 0
> 6960 A3 137 ASP CA CT1 0.070000 12.0110 0
> 6961 A3 137 ASP HA HB1 0.090000 1.00800 0
> 6962 A3 137 ASP CB CT2 -0.210000 12.0110 0
> 6963 A3 137 ASP HB1 HA2 0.090000 1.00800 0
> 6964 A3 137 ASP HB2 HA2 0.090000 1.00800 0
> 6965 A3 137 ASP CG CD 0.750000 12.0110 0
> 6966 A3 137 ASP OD1 OB -0.550000 15.9994 0
> 6967 A3 137 ASP OD2 OH1 -0.610000 15.9994 0
> 6968 A3 137 ASP HD2 H 0.440000 1.00800 0
> 6969 A3 137 ASP C C 0.510000 12.0110 0
> 6970 A3 137 ASP O O -0.510000 15.9994 0
>
> PDB -
> ATOM 6958 N ASP X 137 83.530 3.953 -57.743 1.00 0.00 A3
> ATOM 6959 HN ASP X 137 84.010 4.446 -58.465 1.00 0.00 A3
> ATOM 6960 CA ASP X 137 84.142 3.929 -56.431 1.00 0.00 A3
> ATOM 6961 HA ASP X 137 84.144 2.906 -56.087 1.00 0.00 A3
> ATOM 6962 CB ASP X 137 85.559 4.470 -56.466 1.00 0.00 A3
> ATOM 6963 HB1 ASP X 137 85.928 4.585 -55.425 1.00 0.00 A3
> ATOM 6964 HB2 ASP X 137 85.554 5.463 -56.964 1.00 0.00 A3
> ATOM 6965 CG ASP X 137 86.499 3.570 -57.209 1.00 0.00 A3
> ATOM 6966 OD1 ASP X 137 87.627 4.012 -57.493 1.00 0.00 A3
> ATOM 6967 OD2 ASP X 137 86.117 2.416 -57.490 1.00 0.00 A3
> ATOM 6968 HD2 ASP X 137 86.757 1.882 -57.966 1.00 0.00 A3
> ATOM 6969 C ASP X 137 83.285 4.762 -55.513 1.00 0.00 A3
> ATOM 6970 O ASP X 137 83.301 4.588 -54.306 1.00 0.00 A3
>
>
>
> What is the issue that is happening here? I appreciate the help.
>
>
>
> --
>
> Daniel Guion (he/him)
>
> Graduate Student in Biochemistry, Molecular and Structural Biology
>
> Department of Chemistry and Biochemistry
>
> University of California, Los Angeles
>
> danielguion_at_g.ucla.edu
> 

-- 
Daniel Guion (he/him)
Graduate Student in Biochemistry, Molecular and Structural Biology
Department of Chemistry and Biochemistry
University of California, Los Angeles
danielguion_at_g.ucla.edu