So, you have used g_energy function I suppose. Most of the options are box size, temperature, etc. that you do not want.  You want options from 47.

There are four options for each sets. two for LJ and two for electrostatic for each pair.

Lets take for example protein-LIG pair.

51 is the LJ interaction between them. while 53 is the LJ between atoms that are away by 4 covalent bond. I think they are not bond between these two molecules. So the output for 53  should  be zero.  Also, 50 is the electrostatic energy between them while 52 is between 4 bond away atoms  and again for two separate molecules should be zero. 

So, I think you need 51 and 50. 

Thank you very much. As you said 53 became zero. Do you have any other recommendations for further analysis between the receptor and ligand, which can be useful in publication?

Bests.

you are welcome, you can do hydrogen bond analysis. (in gromacs g_hbond)

Best regards,

Weria 

I did the same for my simulation. Between a small molecule and a DNA fragment.

However, for any possibilities of DNA-LIG interaction the vlaor is zero. Does this mean that there was no interaction during the entire simulation?

A number of molecular mechanics force fields apply the Lennard-Jones function for describing the interaction energy between non-bonded neutral atoms.

A majority of these force fields adopt the Lennard-Jones(12-6) function, followed by the Lennard-Jones(9-6)function. Less common are the Lennard-Jones(12-10), which is available in AMBER as an option in addition to the usual Lennard-Jones(12-6) function, and the buffered Lennard-Jones(14-7) function or simply LJ-14.

The r−12 term, which is the repulsive term, describes Pauli repulsion at short ranges (SR) due to overlapping electron orbitals, and the r−6 term, which is the attractive long-range term, describes attraction at long ranges (LR) (van der Waals force, or dispersion force).

LJ- LR is not given in option so, LJ-SR:Protein-LIG option can be chosen for the VDW interaction, while for electrostatic interaction coul-SR can be chosen.

Hello everyone,

Do you know how to decompose the energy into residue-residue interaction energies, showing electrostatic and van der Waals interactions independently for each reside by using the Gromacs?

Best wishes,

Feng

Use use energy group and define a group for each part that you are interested. If you have already done the simulation, then use rerun option that is much cheaper.

Hello Weria Pezeshkian,

Thank you for your replay! I have finished the energy minimization step without any solvent and ions. Now I want to know the electrostatic and van der Waals interactions independently for each residue of this single point structure. So do you mean that I can use the "gmx energy" with the "index" file to calculate it? I tried to add the index file to the "gmx energy" run, but it remind me: " Unknown command-line option -n". And then how to use the "rerun" option, does it is the "mdrun" option?

Best wishes,

Feng

if you have done energy minimization, then first and solvent and (if needed ions). Then in the final run, add a line like below to your *.mdp file. In your case, you need to provide an index file with different groups which each contains different residue.

add this to your *.mdp

energygrps = residue1 residue2 ...

then after simulation, just run gmx energy -f system.edr

then you get a huge list of interactions, electrostatic and vdw are separated indeed.

I don't need to do the "mdrun" and the solvent, so can I add the "energygrps=" in my em.mdp file, and then run the "gmx energy" directly? But in the previously, when I analysis the potential energy of the energy minimization step(did not have the "energygrps" term in my em.mdp), the potential energy of this structure is positive, so does it has some problem?

Dear Maryam,

Yes, you can use it. Gromacs allows you to find interaction energy between any pair of groups. However, to do this, you need an index file, inside the index file you have to include residuals that are interesting for you. If you only give protein A,B. What you get is, A-B interaction. if you want more details, you need to define more detailed index file as well.