Reduce PDB doesn't match Topology File atom

User discussions
1

GROMACS version: 2023.2
GROMACS modification: No

I am testing running proteins with mutations through AlphaFold and then passing the PDBs to gromacs to run relaxation.

I originally attempted to run this without reducing the molecule, but that definitely didn’t work.

Next, I reduced the structure using GitHub - rlabduke/reduce: Reduce - tool for adding and correcting hydrogens in PDB files. This adds in the hydrogens excluding those for the N-terminus. I think have run pdb2gmx using all of the different options for the force-fields (and TIP3 when applicable) but all result in the topology file adding 3 hydrogens to the N-term which do not appear in the reduced pdb.

I have tried running this with the -ignh flag but it gives the same results.

Additionally, I have attempted to manually add hydrogens to the N-term for the reduced file but using that to generate the topology doesn’t work and neither does using the before topology with a modified pdb work for grompp.

Any assistance/guidance here would be greatly appreciated.

I have also included the bash script I am running below for additional context.

module load gromacs_gpu

reduce -BUILD testing/unrelaxed_model_1_pred_0.pdb > output_with_hydrogens.pdb

gmx pdb2gmx -f output_with_hydrogens.pdb -o output.gro -p output.top -ignh

gmx grompp -f em.mdp -c output_with_hydrogens.pdb -p output.top -o em.tpr

gmx mdrun -v -deffnm em
2

This is the expected (and correct) behavior. A free amino terminus is predominantly protonated (-NH3+) at pH 7. I don’t know why reduce doesn’t protonate it, but pdb2gmx is doing what is chemically sensible.

3

Thanks @jakemhul,
That is what I was thinking too. Do you happen to have suggestions for other methods of reduction and/or ways to get grompp to work when there is a difference in those 3 atoms?

The only option I have been able to find at the moment would be to create a script that modifies the top file and renumbers all of the atoms after deleting the differing 3. I have already tested modifying the pdb file manually to add in the hydrogens and didn’t seem to do the trick.

4

Why would you delete the N-terminal protons? They are chemically necessary for any physically meaningful molecular mechanics process. Use the coordinates and topology from pdb2gmx and don’t edit them, then you will be consistent and should not face issues.