Issue with pdb2gmx when peptide is capped

[mostafagh98]

Hi all,

I’m encountering an issue when trying to process my peptide with pdb2gmx in GROMACS using the CHARMm36-jul2022.ff force field. Here’s the situation:

1. I created my PDB using AlphaFold 3 and added the N-acetyl and C-amide caps using CHARMM-GUI. Here’s part of the PDB file:
   REMARK GENERATED BY CHARMM-GUI (HTTP://WWW.CHARMM-GUI.ORG) V3.7 ON JAN, 04. 2025. JOB
   REMARK READ PDB, MANIPULATE STRUCTURE IF NEEDED, AND GENERATE TOPOLOGY FILE
   REMARK DATE: 1/ 4/25 4: 4: 8 CREATED BY USER: apache
   ATOM 1 CAY LYS P 1 -11.293 -1.388 -11.244 1.00 0.00 PROA
   ATOM 2 HY1 LYS P 1 -10.901 -1.983 -10.392 1.00 0.00 PROA
   ATOM 3 HY2 LYS P 1 -12.232 -0.883 -10.929 1.00 0.00 PROA
   ATOM 4 HY3 LYS P 1 -10.534 -0.631 -11.536 1.00 0.00 PROA
   ATOM 5 CY LYS P 1 -11.570 -2.293 -12.395 1.00 0.00 PROA
   ATOM 6 OY LYS P 1 -11.897 -3.465 -12.216 1.00 0.00 PROA
   ATOM 7 N LYS P 1 -11.443 -1.758 -13.623 1.00 87.90 PROA
   ATOM 8 HN LYS P 1 -11.178 -0.809 -13.777 1.00 0.00 PROA
   …
   ATOM 554 N CYS P 30 15.841 2.920 8.144 1.00 90.41 PROA
   ATOM 555 HN CYS P 30 15.358 3.654 7.673 1.00 0.00 PROA
   ATOM 556 CA CYS P 30 15.016 1.826 8.664 1.00 88.72 PROA
   ATOM 557 HA CYS P 30 13.979 2.119 8.595 1.00 0.00 PROA
   ATOM 558 CB CYS P 30 15.273 0.567 7.848 1.00 81.91 PROA
   ATOM 559 HB1 CYS P 30 16.250 0.135 8.152 1.00 0.00 PROA
   ATOM 560 HB2 CYS P 30 15.332 0.841 6.773 1.00 0.00 PROA
   ATOM 561 SG CYS P 30 13.966 -0.648 8.096 1.00 72.34 PROA
   ATOM 562 HG1 CYS P 30 14.306 -0.930 9.345 1.00 0.00 PROA
   ATOM 563 C CYS P 30 15.324 1.554 10.140 1.00 88.13 PROA
   ATOM 564 O CYS P 30 15.604 0.427 10.508 1.00 81.80 PROA
   ATOM 565 NT CYS P 30 15.264 2.623 10.979 1.00 0.00 PROA
   ATOM 566 HT1 CYS P 30 15.456 2.504 11.954 1.00 0.00 PROA
   ATOM 567 HT2 CYS P 30 15.029 3.526 10.622 1.00 0.00 PROA
   TER 568 CYS 30
   END

2. When I use -ter and select “none” for both the N- and C-terminals, I get the following error: Fatal error: There is a dangling bond at least one of the terminal ends. Fix your coordinate file, add a new terminal database entry (.tdb), or select the proper existing terminal entry.

3. When I select “none” for the N-terminal and “CT2” for the C-terminal, I get this error instead: Fatal error: Atom CAY in residue LYS 1 was not found in rtp entry LYS with 22 atoms while sorting atoms.

Has anyone encountered a similar issue, or could you suggest a solution?

Thanks in advance!

[jalemkul]

CHARMM applies patches by retaining the parent amino acid residue name, so an acetylated N-terminal lysine will still be called LYS. GROMACS cannot handle this case, and requires a distinct residue called ACE that contains the acetylation. The same is true of the C-terminus. You should not have to regenerate a topology after using CHARMM-GUI; it provides you with GROMACS inputs, including the topology provided you selected that option for output.

[mostafagh98]

Thank you so much for your response
I used the PDB Reader and Manipulator option in CHARMM-GUI, and there is no option available for generating GROMACS input files. If possible, could you kindly suggest an alternative solution?

Furthermore, I am studying peptide/DNA interactions as well as DNA/membrane interactions using GROMACS, and I would greatly value your advice on a specific aspect of my simulations. My DNA structure consists of two strands, and I am uncertain about the best way to handle this in my setup.

Would it be more appropriate to define the DNA as a single molecule in the topology file and use tools like pdb2gmx -merge interactive to merge the strands together, or to treat the DNA strands as separate chains?
I want to ensure that I handle periodic boundary conditions properly and avoid any broken molecule appearances in the trajectory.

[jalemkul]

@mostafagh98

I used the PDB Reader and Manipulator option in CHARMM-GUI, and there is no option available for generating GROMACS input files. If possible, could you kindly suggest an alternative solution?

If you build the whole system (including solvation and ions) through step5, you can download everything in GROMACS format. This is the easiest method.

@mostafagh98

Furthermore, I am studying peptide/DNA interactions as well as DNA/membrane interactions using GROMACS, and I would greatly value your advice on a specific aspect of my simulations. My DNA structure consists of two strands, and I am uncertain about the best way to handle this in my setup.

pdb2gmx can handle multiple chains natively. You don’t have to do anything special.

@mostafagh98

Would it be more appropriate to define the DNA as a single molecule in the topology file and use tools like pdb2gmx -merge interactive to merge the strands together, or to treat the DNA strands as separate chains?
I want to ensure that I handle periodic boundary conditions properly and avoid any broken molecule appearances in the trajectory.

Topological merging has no impact on PBC so there’s no need to do this.

[mostafagh98]

Thank you for investing your time