I have converted the itp of my gutathione to rtp format and copied the pdb to my strucuture along with using edit conf to renumber the file with a repacement residue. When I try to convert the pdb I recieve a error
Fatal error:
Residue 175 named GSH of a molecule in the input file was mapped
to an entry in the topology database, but the atom C used in
that entry is not found in the input file. Perhaps your atom
and/or residue naming needs to be fixed.
I checked and residue 175 is a TYR and I cant find any issue with my carbons besides the segmented structure.
The problem is with the GSH residue; the next residue in the sequence expects it to have a C atom to create bonds (e.g. -C in the TYR .rtp entry) and apparently it does not, so pdb2gmx cannot build the expected bonded structure.
OK. I went back and realized the source of my issue was the lack of labeling or missing labels when I convert my pdb to mol2 for cgenff. This issue would result in a str with atom names I dont recognize and cant configure into a proper rtp.
I have since started over confirming my atomnames
I added hydrogens to my residue amine group to satisfy cgenff
.attype warning: monovalent nitrogen not supported;skipped molecule
do I need to also write the [ bonds] ? I started to convert them from the itp but they dont make sense in that there are repeats and connections not present in the pdb.
I dont fully understand how to read or write hdb so I based it off cys since the backbone matches
GSH 4
1 1 HN N CA -C
1 5 HA CA N CB C
2 6 HB CB CA SG
1 2 HG1 SG CB HB1
gmx pdb2gmx -f db_2dwp_gsh_.pdb -o 2dwp_gsh.gro
Program: gmx pdb2gmx, version 2022
Source file: src/gromacs/gmxpreprocess/resall.cpp (line 469)
Fatal error:
in .rtp file in residue moleculetype at line:
GSH 3
This warning suggests something intrinsically wrong with your .mol2 file that should be rectified before even thinking about proceeding. It means the bonded structure is wrong.
I don’t understand this. You should never be making manual edits to coordinates or topologies at this point.
At the bare minimum, yes. Otherwise pdb2gmx has no idea how the molecule is connected and you’ll just get a bunch of atoms in space that aren’t actually bonded in any way. You’ll also need [impropers] if there are any to assign.
I don’t see where this is coming from based on what you’ve provided, but there’s something syntactically wrong with the residue definition.
“attype warning: monovalent nitrogen not supported;skipped molecule”
this was a warning I resolved by adding hydrogens to my nitrogen attached to my CA.
I figured this was because of CGENFF’s intended use on ligands rather than individual members of a polypeptide.
I was trying to match the other rtp files while satisfying CGENFF
I will retry with a new model, the nitrogen in its correct trivalent form bonded to the Calpha and a hydrogen
It concerns me that the resulting mol2 file doesnt display the double bonded oxygens
also can I use atom types to write the bonds of the rtp?
here is the resulting rtp with bonds
`[ GSH ]
;Gutathionylated cysteine 3
[ bonds ]
N CA ; NG321 CG311
N H12 ; CG311 CG321
CA C ; CG311 CG321
CA CB ; CG311 HGP1
CA H17 ; CG321 OG312
C O ; CG321 HGA2
C H23 ; CG321 HGA1
C H16 ; CG321 SG301
CB SG ; CG321 HGA2
CB H09 ; CG321 HGA2
CB H10 ; SG301 SG301
SG S01 ; CG321 SG301
C01 S01 ; CG321 CG311
C01 C02 ; CG321 HGA2
C01 H01 ; CG321 HGA2
C01 H03 ; NG2S1 CG311
N01 C02 ; NG2S1 CG2O1
N01 C03 ; NG2S1 HGP1
N01 H05 ; OG2D1 CG2O1
O01 C03 ; CG311 CG2O1
C02 C05 ; CG311 HGA1
C02 H04 ; NG2S1 CG2O1
N02 C05 ; NG2S1 CG321
N02 C06 ; NG2S1 HGP1
N02 H11 ; OG2D1 CG2O1
O02 C05 ; CG2O1 CG321
C03 C08 ; NG321 CG311
C03 H24 ; NG321 HGA1
N03 C10 ; NG321 HGPAM2
N03 H20 ; OG2D1 CG2O2
N03 H22 ; OG311 CG2O2
O03 C07 ; OG311 HGP1
O04 C07 ; OG2D1 CG2O2
O04 H18 ; CG321 CG2O2
C05 H25 ; CG321 HGA2
O05 C11 ; CG321 HGA2
C06 C07 ; OG311 CG2O2
C06 H13 ; OG311 HGPAM2
C06 H14 ; CG321 CG321
O06 C11 ; CG321 HGA2
O06 H19 ; CG321 HGA2
C07 H26 ; CG321 CG311
C08 C09 ; CG321 HGA2
C08 H07 ; CG321 HGA2
C08 H08 ; CG311 CG2O2
C09 C10 ; CG311 HGPAM2
C09 H06
C09 H15
C10 C11
C10 H21
C11 H27 `
I dont know if the charge#s are correct. I copied them from the str, but they dont match the patterns of the residues in the charmmff rtp.
I ran through pdb2gmx
-------------------------------------------------------
Program: gmx pdb2gmx, version 2022
Source file: src/gromacs/gmxpreprocess/add_par.cpp (line 165)
Fatal error:
Atom HN not found in rtp database in residue GSH, it looks a bit like H12
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------
I took this as gmx expecting an HN for my backbone amine with the name HN so I renamed my H12 HN
and re ran pdb2gmx
Opening force field file ./charmm36-jul2021.ff/aminoacids.arn
Checking for duplicate atoms....
Generating any missing hydrogen atoms and/or adding termini.
Segmentation fault (core dumped)
Program: gmx pdb2gmx, version 2022
Source file: src/gromacs/gmxpreprocess/add_par.cpp (line 165)
Fatal error:
Atom HN not found in rtp database in residue GSH, it looks a bit like H12
it seems like there is a process of confirming residues by locating the expected HN which my residue lacks.
I changed H12 to HN in my rtp and pdb and reran
-------------------------------------------------------
Program: gmx pdb2gmx, version 2022
Source file: src/gromacs/gmxpreprocess/pdb2gmx.cpp (line 873)
Fatal error:
Atom H in residue GSH 205 was not found in rtp entry GSH with 52 atoms
while sorting atoms.
For a hydrogen, this can be a different protonation state, or it
might have had a different number in the PDB file and was rebuilt
(it might for instance have been H3, and we only expected H1 & H2).
Note that hydrogens might have been added to the entry for the N-terminus.
Remove this hydrogen or choose a different protonation state to solve it.
Option -ignh will ignore all hydrogens in the input.
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------
i added ingh option with the HN substitution and
-------------------------------------------------------
Program: gmx pdb2gmx, version 2022
Source file: src/gromacs/gmxpreprocess/add_par.cpp (line 165)
Fatal error:
Atom HA not found in rtp database in residue GSH, it looks a bit like HN
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------
How can I resolve my HA HN and any other hydrogen bonding issue? are there specifed names for the backbone atoms of all residues that I need to adhere to when designing my GSH?
Your molecule is essentially a polypeptide with an unusual linkage. That’s all glutathione is. I would name everything like a standard amino acid, using all the same nomenclature. Honestly, you can assign all the atom types and charges directly from existing residues rather than going through all this mess.
How would you do this? is there an easier way than transcribing all the cys and gly and Glu bonds first then combining them in no specific order and pulling the s-s from my gsh.
Ive never written an rtp file and I dont know whats wrong with my current one.
Im currently running pdb2gmx and correcting every hydrogen name it spits out as “not found in rtp”
-------------------------------------------------------
Program: gmx pdb2gmx, version 2022
Source file: src/gromacs/gmxpreprocess/add_par.cpp (line 165)
Fatal error:
Atom HG1 not found in rtp database in residue GSH, it looks a bit like HN
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------
why is gmx trying to build cys from GSH HG1 is the hydrogen that caps the sulfur of cys but doesnt exist in my GSH
Because your .hdb entry tells it to. But that entry is poorly constructed and shouldn’t be used because the atom names are not correct and it is incomplete. Pattern the Cys portion of GSH after the disulfide form of Cys in the force field. I would suggest leaving the other Cys moiety of the disulfide off of this molecule so it is treated as a normal part of the protein, and add on the GSH as a new entry in specbond.dat. It will make your life far easier.
I pushed it through pdb2gmx -ingh as to follow the hbd i wrote instead and it looks properly assembled.
First off thank you for your help!
my only question is how would you confirm the assembly?
since completing my .gro i also added to the ffbonded.itp based on the gsh.prm and itp data.
I have run into errors /w trying to add ion’s with grompp
ERROR 1 [file topol.top, line 36837]:
No default U-B types
ERROR 2 [file topol.top, line 36838]:
No default U-B types
ERROR 3 [file topol.top, line 51480]:
No default Proper Dih. types
ERROR 4 [file topol.top, line 51481]:
No default Proper Dih. types
ERROR 5 [file topol.top, line 51482]:
No default Proper Dih. types
ERROR 6 [file topol.top, line 51483]:
No default Proper Dih. types
ERROR 7 [file topol.top, line 51484]:
No default Proper Dih. types
ERROR 8 [file topol.top, line 51485]:
No default Proper Dih. types
ERROR 9 [file topol.top, line 51486]:
No default Proper Dih. types
ERROR 10 [file topol.top, line 63449]:
Unknown cmap torsion between atoms 2772 2814 2816 2824 2826
There was 1 note
-------------------------------------------------------
Program: gmx grompp, version 2022
Source file: src/gromacs/gmxpreprocess/toppush.cpp (line 1715)
Fatal error:
There were 10 errors in input file(s)
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------
I added my dihedrals from the created prm ( charmm2gmx) along with my bonds. How would you address these?
also is there a helpful guide for adding to the ffnonbonded.itp?
I looked over another discussion
and reading the info of the errors the top is listing contacts between my new residue and the backbone atomistic contacts that are missing description that should be covered in the dih types of the ffbonded.itp
do i use similar entries to describe these dihedrals?
for example
ERROR 4 [file topol.top, line 51483]:
No default Proper Dih. types
I see in the topology file ln 51483
2748 2766 2768 2770 9
which refers to the solv.gro
204LYS N 2746 10.735 7.696 4.086
204LYS HZ2 2764 10.981 7.390 4.350
204LYS C 2766 10.837 7.764 3.868
205GSH CA 2770 10.763 7.793 3.652
so i would look up another instance where lys is next to cys and add it to the ffbonded.itp?
otherwise build a 3 peptide chain of the residues surrounding gsh and push it through cgen to copy the missing dihedrals?
Where did you add them? What are the lines you added? It is sufficient to #include a topology file that lists any new bonded parameters rather than necessarily hack ffbonded.itp.
Thank you. I thought that wouldnt work for residues
the ligand topology from charmm doesnt match the rtp I wrote. I dont think I can just copy it over to my rtp because the backbone doesnt match
would I be able to just add the itp and prm for the r-group since the bonded and nonbonded parameters of the backbone match cys (and should be already included) or do the parameters need to be specified.
should I rewrite the rtp (mainly the r-group, keeping the cys-based backbone) and rerun pdb2gmx with the renamed retyped rtf and hbd (based on the charmmgui topology)?
or
could I just make the addition to the R-group a separate residue in the rtf and the hbd, break up the construction into 2 parts?
Could I just change my cys to a disulfide cys2 and add my gutathione as a ligand?
how would I describe this in my rtp and hbd? could I just write gutathione with a incomplete sulfur?
ultimately i dont understand how to add my new residue topology parameters and coordninates to the protein.
Also should I create a new residue to add in place of cys to insert into the protein or just a new R group to add as a covalent attachment ?
currently I understand my errors are due to my bonded and nonbonded information missing from my top file in the form of prm and itp files
I went forward and wrote a new pdb with cys2 in place of ptm site
i followed your protein ligand instructions in designing my gutathione as a ligand and using cgenff to parameterize it.
after parameterization I removed the backbone atoms and converted it to a .gro
i pasted it in to make my complex.gro
Im having trouble customizing the itp to only represent the R group of my original structure.
also how do i write this as covalently attached to cys2?
I rewrote my specbonds.dat but how do i call for that to be applied in this .top
ERROR 2 [file gsh3.itp, line 52]:
Atom index (37) in bonds out of bounds (1-34).
This probably means that you have inserted topology section "bonds"
in a part belonging to a different molecule than you intended to.
In that case move the "bonds" section to the right molecule.
currently I am correcting the bonds to only connect the remaining r group atoms
Correct me if I’m wrong. I currently see 2 ways to build my structure. The recommended route of :
building my alternative residue pdb,
Formatting to determine the topology (cgenff),
Then formatting the structure based on the topologies already provided in the gmx rtp (identify the names and types then replace the names types and bond information of the itp to match a traditional residue, for me I could see this working by replacing the backbone of gsh with atoms matching cys)
paste the formatted itp in the rtp
write up an hbd for new residue
rewrite the .gro adding the coordinates of the gsh r group to the cys and renaming it GSH
the other way which I am currently working is to
change cys to cys2 with an available sulfur
treat the R-group as a ligand
convert the pdb to gro add it to the protein .gro
build the pdb, format it to push through cgenff then remove the backbone
change the sulfur to something indicating its attachment to the cys2 sulfur atom type sm in the ligand itp
edit the $gmx top specbonds.dat to allow the bond.
also could you help me understand the [pairs] section of the itp
Is there a better way than reading through the itp and looking up each line of atoms bonds and dihedrals to rewrite it and then rewrite it again when the numbers refering to the codes change?
The simplest thing to do here is to simply create an all-in-one residue. Start with the disulfide Cys and build out from the backbone by adding the γ-Glu-Cys(disulfide form)-Gly. Pretty much all the bonded connectivity and associated .hdb information comes directly from the existing residues. The only difference is the γ-peptide bond, but you can take all the charges and atom types from the backbone of existing residues. There may be a missing dihedral or two, but those are simple to introduce by analogy.
