Small questions about covalent bound ligand

jxdlneto · January 3, 2025, 2:10am

GROMACS version: 2021.3
GROMACS modification: No

Dear users,
I know there are many questions here about the MD of protein/covalent-bound ligands… but I need to do more about some details… I’d appreciate it if anyone could help me!

I obtained the itp/prm files of the cys-ligand properly hydrogenated and included the information in the aminoacids.rtp. However, when I was creating an entry in the aminoacids.hdb file I didn’t understand the organization:

CYS        4
1       1       HN      N       CA      -C     
1       5       HA      CA      C       CB      N      
2       6       HB      CB      CA      SG     
1       2       HG1     SG      CB      CA

What are these numbers in columns 1 and 2? Is column 4 the atom to which hydrogen is bonded while the others are the atoms bonded to it?

The other questions are about the ffbonded.itp.
In the manual, it is shown that " If you require any new bonded types, add them to ffbonded.itp". Since my entire ligand plus the covalent bond are “new bonded types” (at least different from protein), should I include everything in the prm file in the ffbonded.itp (I think the prm is closer to the information that the itp need) or just the information about the covalent bond?

One last question about this itp file. The str file generated by Cgenff does not have the b0 and kb of the bond (cystein)S-(ligand)C (but showed angles and dihedral). How can I get these values? I believe b0 could be the distance in the PDB file, but I think the kb isn’t so simple to find…

Thanks for your help and time

jalemkul · January 3, 2025, 1:17pm

See the reference manual: pdb2gmx input files - GROMACS 2024.4 documentation

The parameters you see listed in the .str file are ones that are new and otherwise not already present in the force field. Every other interaction is already present in the parent force field, meaning you do not need to do anything.

jxdlneto · January 3, 2025, 8:29pm

Thank you @jalemkul.

Ok. So, in the example of Cysteine residue, the second line means 1 hydrogen planar HN added, the fourth line added 2 tetrahedral HBs, etc, is that right? One more question. My new Cys-lig residue has 30 hydrogen atoms, must I create 30 lines or (for example) HB2 and HB3 are bonded to the same CB atom, can I add a line 2 (atoms) 6 (two tetrahedral hydrogens) HB CB .... (H1 and H3 are bounded to the same C5: 2 6 H C5...)?

OK. I have nothing to do with the S-C… but I need to introduce all other parameters present in the .str (.prm) file in the ffbonded.itp, is that right?

jalemkul · January 5, 2025, 4:03am

Not necessarily 30 lines, but note that GROMACS will start numbering from 1, so you will not generate HB2 and HB3, you will generate HB1 and HB2. If you have atoms H1 and H3 bonded to a single atom, those have to be specified with individual lines, not 2 at once (because you’ll get H1 and H2, not H3).

Yes.

jxdlneto · January 5, 2025, 4:16am

Hum… I don’t know if I understood it…
So do I have to change the numbering in the .rtp file (e.g. HB2 to HB1) or doesn’t it matter (and it is only something for GROMACS)?

jalemkul · January 5, 2025, 7:45pm

You will make your life easier if you adopt the numbering convention in GROMACS. Here is an example.

2 6 HB CB CA CG

This generates 2 HB atoms (HB1 and HB2) in a tetrahedral configuration around CB. If you want to generate two atoms named H1 and H3 bonded to CB in the same geometry, you would have to do something like:

1       6       H1      CB      CA      CG
1       6       H3      CB      CG      CA

Note the inversion of the last two control atoms to invert the stereochemistry of the added H, otherwise they will be added on top of one another and you’ll have an infinite force during minimization.

jxdlneto · January 7, 2025, 9:37am

Dear, @jalemkul

Everything looks like to be working fine up to now…
I want to ask 2 more questions:

What should I include in the aminoacids.rtp file [ improper ] and [ cmap ]? Could I maintain the normal lines that were in the DCYS plus the lines in .prm file?

  [ impropers ]
        N    -C    CA    H
        C    CA    +N     O
        C3    C2    N5    SG
        C21    C22    N4    O4
  [ cmap ]
       -C     N    CA     C    +N

How can I include the peptide bond of the new residue (Cys+lig) with the rest of the protein? Is it only include the line C +N, as follows?

  [ bonds ]
       N    CA
       N    H
       C    +N

jalemkul · January 7, 2025, 5:48pm

Yes.

jxdlneto · January 9, 2025, 6:57pm

OK

To include the .prm information in the ffbonded.ipt, is it necessary to change the type to atom’s “name”? For example:

CG2D1O    CG311     1   0.15092000    287859.20
to 
C3    C2     1   0.15092000    287859.20

Besides, since C1, N1, etc are not atoms in the residues by default, should I include these atoms in the ffnonbonded.itp? If yes, how can I get the sigma and epsilon values?

jalemkul · January 10, 2025, 2:25pm

No. Parameters are assigned by type.

jxdlneto · January 10, 2025, 9:45pm

OK.
I test here and I got 3 warning/error messages after running grompp:

It shows the following message for all the new types (except hydrogen atoms). What should I do? Should I rename the types (in ffbonded and aminoacids.rtp) or search for the types in the ffbonded and see if the bonds/angles/dihedrals are already present (deleting the new ones)?

WARNING 1 [file ffnonbonded.itp, line 449]:
  Atomtype NG321 was defined previously (e.g. in the forcefield files), and
  has now been defined again. This could happen e.g. if you would use a
  self-contained molecule .itp file that duplicates or replaces the
  contents of the standard force-field files. You should check the contents
  of your files and remove such repetition. If you know you should override
  the previous definition, then you could choose to suppress this warning
  with -maxwarn.

It showed this error. Reading the ffnonbonded.itp, the line shows OG2D1 8 15.999000 0.000 A and does not have the sigma and epsilon values. Is there a way to fix it? OG2D1 is the type for all C=O oxygen atoms

 WARNING 4 [file ffnonbonded.itp, line 452]:
  Too few parameters on line (source file
  /home/xavie/gromacs-2021.3/src/gromacs/gmxpreprocess/toppush.cpp, line
  343)

It shows the following error for all the bonds, angles, and dihedrals related to the atoms in the peptide bond. How can I fix it (I added the line C +N in [ bonds ] of the aminoacids.rtp file)?

ERROR 1 [file topol.top, line 7344]:
  No default Bond types

ERROR 3 [file topol.top, line 26428]:
  No default U-B types

ERROR 11 [file topol.top, line 36974]:
  No default Proper Dih. types

ERROR 31 [file topol.top, line 44106]:
  No default Improper Dih. types

ERROR 33 [file topol.top, line 44702]:
  Unknown cmap torsion between atoms 2221 2223 2225 2232 2234

jalemkul · January 13, 2025, 2:04pm

You should never have to mess with anything in [atomtypes] in the force field files. The duplicate definition means you already have the N type that you need so you shouldn’t be adding it again. A misformatted line means you added an incomplete line.

As for the missing parameters, I can’t say anything about them without knowing what those parameters are and if they were added to your original .str file as new parameters. The CMAP torsions won’t be; you’ll have to add those by analogy because the backbone doesn’t differ among most amino acids (only Gly and Pro are different).

jxdlneto · January 14, 2025, 2:39am

How can I do this? I ask because I didn’t change the ffnonbonded.itp. So, I didn’t include these atomtypes. The forcefield files I’ve changed were aminoacids.hbd, aminoacids.rtp and ffbonded.itp (I copied all lines in .prm file and pasted them here).

You said, change the -N from NG321 to NH1, CA from CG311 to CTD1, etc?

jalemkul · January 15, 2025, 2:54am

Something isn’t adding up but without access to your files, I can’t tell what. The complaint is about multiply defined atom types so I can’t see how else you would be getting that unless you have an #include statement somewhere else that’s causing the problem.

To solve the CMAP issue, you would just copy the standard CMAP entry for the canonical protein atom types and use the CGenFF atom types.

jxdlneto · January 17, 2025, 5:18pm

Which files must I share?

jalemkul · January 17, 2025, 8:00pm

Anything that you’re editing that is producing an error message.

jxdlneto · January 20, 2025, 9:50pm

Dear @jalemkul
I changed the [ atomtypes ] of the cysteine residue in the cys-lig (CSL) to those in the DCYS, maintaining the CGenFF ones in the ligand. Instead of 34 errors related to the atoms in the cys backbone, it showed 10 errors:

WARNING 1 [file ffbonded.itp, line 6991]:
  Bondtype U-B was defined previously (e.g. in the forcefield files), and
  has now been defined again. This could happen e.g. if you would use a
  self-contained molecule .itp file that duplicates or replaces the
  contents of the standard force-field files. You should check the contents
  of your files and remove such repetition. If you know you should override
  the previous definition, then you could choose to suppress this warning
  with -maxwarn.

  old:                                          110 365.682 0 0 110 365.682 0 0 
  new: NH1     CTD1        C     5   107.000000   418.400000   0.00000000         0.00


ERROR 1 [file ffbonded.itp, line 21554]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


ERROR 2 [file ffbonded.itp, line 21594]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


ERROR 3 [file ffbonded.itp, line 21662]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


ERROR 4 [file ffbonded.itp, line 21666]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


Generated 167799 of the 167910 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 1

Generated 117432 of the 167910 1-4 parameter combinations

ERROR 5 [file topol.top, line 7355]:
  No default Bond types


ERROR 6 [file topol.top, line 26465]:
  No default U-B types


ERROR 7 [file topol.top, line 37012]:
  No default Proper Dih. types


ERROR 8 [file topol.top, line 37013]:
  No default Proper Dih. types


ERROR 9 [file topol.top, line 37014]:
  No default Proper Dih. types


ERROR 10 [file topol.top, line 37015]:
  No default Proper Dih. types

Is there a problem doing this?
The errors 1-4 are related to the atoms in the cysteine backbone. Could I solve it by putting the correspondent line below the actual line? For example:
NH1 CTD1 C O … mult 4
NH1 CTD1 C O … mult 5 (the new one)
The errors 5-10 are related to the new bond Cys(S)-Lig(C).

Here are the files modified and showing errror

jxdlneto · January 21, 2025, 10:31pm

I made some more changes in the files, and I got less error messages:

WARNING 1 [file ffbonded.itp, line 6991]:
  Bondtype U-B was defined previously (e.g. in the forcefield files), and
  has now been defined again. This could happen e.g. if you would use a
  self-contained molecule .itp file that duplicates or replaces the
  contents of the standard force-field files. You should check the contents
  of your files and remove such repetition. If you know you should override
  the previous definition, then you could choose to suppress this warning
  with -maxwarn.

  old:                                          110 365.682 0 0 110 365.682 0 0 
  new: NH1     CTD1        C     5   107.000000   418.400000   0.00000000         0.00


ERROR 1 [file ffbonded.itp, line 21662]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


ERROR 2 [file ffbonded.itp, line 21666]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


Generated 167799 of the 167910 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 1

Generated 117432 of the 167910 1-4 parameter combinations

ERROR 3 [file topol.top, line 7355]:
  No default Bond types


ERROR 4 [file topol.top, line 26465]:
  No default U-B types

but the questions continue…

Is there a problem changing the [ atomtypes ] to cysteine types?
Can I solve the errors 1 and 2 by putting the correspondent line below the actual line? For example:
NH1 CTD1 C O … mult 4
NH1 CTD1 C O … mult 5 (the new one)
How can I solve the errors 3-4 since they are related to the new bond Cys(S)-Lig(C)?
The files are in the same shared folder:
Here are the modified files

jxdlneto · January 25, 2025, 10:40pm

Dear,

Let me review what I have done so far to make clear the details:

I got the PDB file, changed the position of the ligand to be “part” of the cysteine residue (now called CSL), and renumbered the atoms. So, I completed the hydrogen atoms only in the CSL residue (capping the N and C terminals with hydrogen) and used the CGENFF to get the parameters.

Then I deleted the lines of the hydrogen atoms used to cap the N-C terminals in the .str file and used the lines of [atom] and [bonds] of the .itp file to include these parameters of the new CSL residue in the aminoacids.rtp. I also included the new residue in the residuetype.dat (CSL Protein) and the hydrogens in the aminoacids.hdb. Finally, I added the lines of the .prm file (bonds, angles and dihedrals) in the ffbonded.itp at the end of the par_all36_cgenff.prm parameters (bonds, angles, and dihedrals).

I’m stuck at this point… If it is possible to help me continue, I’ll greatly appreciate it. When I tried to use the [ atomtypes ] obtained from CGENFF, the grompp shows more than 30 errors, most of them related to the peptide bond, but when I changed the [ atomtypes ] for those in the CYS residue, it showed 4 errors (two of then are related to the new (cys)S-C(lig) bond.

The files of the first test

**Erros in the first test**
ERROR 1 [file topol.top, line 7344]:
  No default Bond types

ERROR 3 [file topol.top, line 26428]:
  No default U-B types

ERROR 11 [file topol.top, line 36974]:
  No default Proper Dih. types

ERROR 31 [file topol.top, line 44106]:
  No default Improper Dih. types

ERROR 33 [file topol.top, line 44702]:
  Unknown cmap torsion between atoms 2221 2223 2225 2232 2234

The files of the second test

**Errors in the second test:**
WARNING 1 [file ffbonded.itp, line 6991]:
  Bondtype U-B was defined previously (e.g. in the forcefield files), and
  has now been defined again. This could happen e.g. if you would use a
  self-contained molecule .itp file that duplicates or replaces the
  contents of the standard force-field files. You should check the contents
  of your files and remove such repetition. If you know you should override
  the previous definition, then you could choose to suppress this warning
  with -maxwarn.

  old:                                          110 365.682 0 0 110 365.682 0 0 
  new: NH1     CTD1        C     5   107.000000   418.400000   0.00000000         0.00


ERROR 1 [file ffbonded.itp, line 21662]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


ERROR 2 [file ffbonded.itp, line 21666]:
  Encountered a second block of parameters for dihedral type 9 for the same
  atoms, with either different parameters and/or the first block has
  multiple lines. This is not supported.


Generated 167799 of the 167910 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 1

Generated 117432 of the 167910 1-4 parameter combinations

ERROR 3 [file topol.top, line 7355]:
  No default Bond types


ERROR 4 [file topol.top, line 26465]:
  No default U-B types

Topic		Replies	Views
Another doubt about covalent bound ligands User discussions	0	40	October 4, 2024
Adding a covalent ligand to a force field User discussions forcefield	18	3885	October 13, 2024
Issue with covalent protein-ligand model/Adding ions step User discussions forcefield , simulation-setup	0	366	October 14, 2023
Gromacs tutorial for covalently bound protein-ligand complex? User discussions gmx-tutorials	0	22	February 9, 2025
Protein/cofactor + ligand User discussions forcefield , simulation-setup	7	1093	May 5, 2021

Small questions about covalent bound ligand

Related topics