Error: No Default LJ-14 Types When Integrating Small Molecule into GAFF+GLYCAM System

User discussions
,
1

GROMACS version: 2022.5
GROMACS modification: Yes (include patch plumed 2.9.1)

Hello GROMACS community,

I am working on a system with a microgel that uses a mixed force field (GAFF + GLYCAM) along with water in a simulation box. I recently modified the system to include a small molecule (ligand), which is set for GAFF. I am trying to integrate this small molecule by calling its ITP file into my existing topology file (.TOP), instead of regenerating the topology file from Amber (using tleap and ACPYPE).

Here’s what I’ve done so far:

I added the following line to my existing topology (topol.top):

#include "oxd.itp"

Updated the [ molecules ] section to include the small molecule:

[ molecules ]
; Compound                nmols
diamantane                1
SOL                       189204
oxd                       39

I used insert-molecules to modify the gro file and so be able include the small molecule’s coordinates.

However, when I run gmx grompp, I receive an error for each atom in the small molecule, such as:

ERROR 143 [file oxd.itp, line 273]: No default LJ-14 types

It appears that the Lennard-Jones 1-4 interaction types for the ligand are not being recognized. Since I’m using a mixed force field system (GAFF for the ligand and GAFF+GLYCAM for the microgel), I’m unsure if this is due to force field compatibility issues, missing parameters, or something else related to the nonbonded interactions.

My questions are:

1-How can I resolve the “No default LJ-14 types” error for my small molecule without going back to tleap (and acpype) to prepare again the entire system?
2-Is this error likely caused by a conflict between GAFF and GLYCAM parameters, and if so, how can I manage these interactions properly in GROMACS?

Any help or advice would be greatly appreciated! Thanks in advance!

2

It seems like there are [ pairs ] entries in “oxd.itp” without matching [ pairtypes ] in the forcefield. So, the parameters are missing.

From where did you get the ligand itp file? ACPYPE usually includes all the necessary parameters. You wouldn’t have to run ACPYPE to regenerate the whole topology file, but you might have to regenerate the ligand itp file. I don’t think you need to use tleap at all for that.

Provided that my above analysis is correct, there is no conflict between the parameters. It’s just that ACPYPE usually generates the parameters required for the input, but not all possible parameters. So, if you add something from another source you have to make sure that all those required parameters are defined.

3

Thank you for your input, Magnus.

I initially generated the ligand ITP file using ACPYPE, starting from a MOL2 file with RESP charges. However, after redoing the process, I still encountered missing r and k values in the [ pairs ] section for the Lennard-Jones 1-4 interactions. It seems like ACPYPE isn’t including these parameters, and as a result, GROMACS is giving the “No default LJ-14 types” error.

I’ve confirmed that the atom types in the ITP file match those from GAFF2, but the required parameters for the 1-4 interactions aren’t being generated by ACPYPE. I understand that I might need to calculate these manually using combination rules from the force field, but I’m unsure how to do that.

Could you explain how I should calculate these missing values for the Lennard-Jones 1-4 interactions? Any additional steps or tools you’d suggest to ensure all parameters are properly included would be really helpful.

Thanks again for your help!

4

In the Amber force fields I can’t even find any [ pairtypes ] sections. So, if there are entries in [ pairs ] that do not have parameters specified you will run into this problem (missing default parameters).

I would not recommend trying to calculate the missing parameters, at least not as a first alternative. I think it would be better to contact the GAFF2 force field developers, or possibly @alanwilter, who is developing ACPYPE, and see if they can help you.