An error occurred while executing the Adding Ions command

User discussions
,
1

GROMACS version: 2020.1
GROMACS modification: No

I found an error message like this, previously I used the pdb2pqr server to change the pH of my protein.pdb with the AMBER force field. I want to do a molecular dynamic simulation with pH variations in protein, with the hope of getting different rmsd stability results for each pH variation. Thanks in advance

gmx grompp, version 2020.1-Ubuntu-2020.1-1
Executable: /usr/bin/gmx
Data prefix: /usr
Working dir: /mnt/c/Users/user/Downloads/coba 3/propka-complex AMBER no2
Command line:
gmx grompp -f ions.mdp -c solv.gro -p topol.top -o ions.tpr

Ignoring obsolete mdp entry ‘title’
Ignoring obsolete mdp entry ‘ns_type’

NOTE 1 [file ions.mdp]:
With Verlet lists the optimal nstlist is >= 10, with GPUs >= 20. Note
that with the Verlet scheme, nstlist has no effect on the accuracy of
your simulation.

Setting the LD random seed to 810795547

ERROR 1 [file kni.prm, line 4]:
Unknown bond_atomtype CG2RC0

There was 1 note

Program: gmx grompp, version 2020.1-Ubuntu-2020.1-1
Source file: src/gromacs/gmxpreprocess/toppush.cpp (line 762)

Fatal error:
There was 1 error in input file(s)

For more information and tips for troubleshooting, please check the GROMACS
website at Errors - Gromacs

2

CG2RC0 is a CGenFF atom type and is provided in the CHARMM36 port. Is this the force field you’re using? Please provide all steps taken up to this point. Missing atom types usually suggest a user trying to mix and match force fields, or use some incompatible version.

3

Thanks for your feedback, I used AMBER 03 & AMBER 94 style fields and the result still shows the same error message as above. For the steps, I followed exactly what is on the Protein-Ligand Complex tutorial web. For protein preparation, I first used the web (https://server.poissonboltzmann.org/pdb2pqr) to change the pH of the protein, and then I analyzed the difference in RMSD and stability of the complex based on variations in pH.
Before I processed it with Gromacs, I changed the protein format from .pqr to .pdb using OpenBabel software.

Here I include the file that I tried to use for the simulation :
https://drive.google.com/drive/folders/1nb4sJyldy0IJ1VI4p_dsn-RJ3kugkHMO?usp=sharing

4

As I suspected, you’re mixing force fields. You can’t use AMBER for the protein and CGenFF for the ligand. Force fields are self-consistent entities; you can’t use AMBER for some part and CHARMM for the other. It’s fundamentally invalid. Either you need to parametrize the ligand for use with AMBER (though the available AMBER protein force fields are quite outdated), or use CHARMM for the protein to work in concert with a CGenFF ligand topology.

5

How do I create a ligand parameter to use in the AMBER force field like creating a CHARMM parameter in the CGenFF web? is it possible without installing Antechamber on my computer?

6

There are lots of tools available to generate AMBER parameters, including RESP ESP charge Derive: R.E.D. version III - R.E.D. III and others. Anything local will require the use of antechamber, AFAIK.

7

Thanks for the advice sir, do I still have to use the command :
python cgenff_charmm2gmx.py JZ4 jz4_fix.mol2 jz4.str charmm36-mar2019.ff
when I create a ligand topology using AMBER ff?

8

No, that conversion script is specifically for CGenFF topologies.

9

I’m sorry sir, I’ve tried the AMBER website that you suggested, but I don’t get the ligand.prm output which is the same as the ligand.prm output on the CGenFF web when I input the ligand.mol2 file

Here I attach the ligand.prm file that I got from the web acpype (AMBER) & CGenFF web :
https://drive.google.com/drive/folders/1oTWFbcsSdSfGpEKGJZYPk2rkKEogevQ0?usp=sharing

10

@aldowanandy I looked at your files, it’s a total confusion.

Your knifix_CHARMM (acpype).prm is in CHARMM format but with AMBER FF. However you want to run with gromacs right? You can only use knifix_CHARMM (acpype).prm if you running a system all built in Amber FF using Charmm programme! You’re mixing programmes with FFs.

If you need help with acpype you need to give me as much information as possible.

Different pHs means just different protonations for some specific amino acids like HIS (HIE, HID, HIP).

I don’t know why you’re using these old AMBER FFs. Knowing almost nothing about your problem I’d say to use AMBER14SB for the protein and GAFF2 for the Ligand.

That said, I don’t have a tutorial for that but you can try to abstract what you need from these two here:

  1. Tutorial Using ACPYPE for GROMACS · alanwilter/acpype Wiki · GitHub
  2. Tutorial NAMD · alanwilter/acpype Wiki · GitHub

In (1) it uses Amber99SB that comes with Gromacs, but it’s old as we said. To use something modern, perhaps (2) is enough for your case (just skip the NAMD part).

Now, how your ligand will change charges in different pHs? Well, that’s up to you to figure out. I presume that https://server.poissonboltzmann.org/pdb2pqr is suppose to help you here.

But, as @jalemkul pointed out, RESP ESP charge Derive: R.E.D. version III - R.E.D. III is the best way to get the charges, yet not the easiest. acpype uses antechamber with bcc to get the charges, which is fair for most cases but I don’t know about yours in different pHs. You probably need to read a bit more in the literature.

Once you get the charges, you can use acpype ... -c user option where your charges are given in a mol2 file.

I hope it helps. Good luck!