Follow the suggestions printed by the program:

1. If there are any close contacts between the water molecule and protein, adjust the position of the water molecule or remove it. 

2. If there are no obvious clashes but you still get the error, reduce the timestep (e.g., 0.01 fs) and lower the temperature (e.g., 5 K).

Try checking the geometry file .pdb/.gro for the coordinates. Review the file for any overlap of molecules or particles sharing the same coordinate.

Another interesting hack is to load the geometry file in VMD. It would warn you about unusual bonds for every close contacts/overlap. Track the particle number and change it accordingly. 

I assume you have already tried reducing the time step.

It seems your system is not equilibrated well. therefore, use the attached .mdp file. run a simulation using it, then perform short NVT. on the next step  you can use output for NPT simulation.  

Thank you everyone. The problem was solved. There was a problem in coordinates. 

how to solve this problem on coordinates?

is *.gro file corrected by which water molecules are deleted? and then to revise the topol.top file?

 how have you solved that problem? can you tell me?

Hey Yang and Zahidul. Sorry to tell you but i dont exactly remember how i solved it. This was way back in july when i was guiding some students. Sorry for not being much of a help

Hey Yang and Zah, there is no standard procedure to tackle this problem. As Jamol. said this is usually a matter of improper equilibration. But there are cases where it is due to problematic coordinates, especially when you want to introduce something new. I suggest to open a new thread with more details so the community can be of more help.

Hi, everyone I have face the same problem,  i had solved this problem discuss  with my senior colleague #Baba Sailu. Ist Energy minimize the complex & edit the minim.mdp file, editing in emsteps = 0.01 replace into 0.02. That's the problem is over out. !!!

Sorry to reply late. Just optimize/energyminimize/clean geometary of protein using Chimera or Discovery studio. It works perfectly

I have met the same problem, I changed emsteps in minim.mdp file, but it doesn't work. I wanna know how to solve the problem.

step 22: Water molecule starting at atom 781343 can not be settled.

Dear Nan, I have found some solutions right here on this forum that helped me on solving this. Both possibilities involve the use of Accelrys Discovery Studio Client software. 1) In the simplest of them, you just open the file with the coordinates and simply rotate the molecule along any axis. Any rotation or translation is enough. Save the coordinates and repeat the process on GROMACS. 2) In the other solution, you open the coordinate file and, using the native optimization tool in Discovery Studio, run about 2 or 3 minimization cycles and save the coordinates. I think it should be enough.

Thank you so much for your help, Daniel. I will have a try.

OK. Please let us know if it works.

Daniel, when you say coordinate file, do you mean .gro file? I'm facing the same problem with my protein.

Dear Gabriela, good morning. I mean the PDB file, which can be generated from the GRO format with OpenBabel or Gromacs (editconf or trjconv). I recommend that you generate the PDB file from the GRO coordinates and proceed as I wrote earlier. Does it help?

Daniel, I am sorry my late reply. I just changed "define" parameter from MDP file to -DFLEXIBLE and my simulation worked well. I really appreciate your quick response! Best regards

I also faced this problem, but i didn't find any proper resolution of my problem, this is problem is due to close contact between protein and water, can be resolve by two way, first way to edit the minim.mdp file, editing in emsteps = 0.01 replace into 0.02 but some times it also failed, second way to solve this by repeating the whole procedure again. Delete every thing except protein, ligand .pdb file and ligand .itp file,again construct the topology, add water and ions. It works without editing minim.mdp file.

Dear, Sir@Vijay Kumar Vishvakarma,@Gabriela de Lima Menezes

First, You have done Energy minimization of your docked complex any of the online server or Discovery Studio or Schrodinger or any other tool. After, that you have prepared Ligand topology file any of the server like-wise PRODRG etc, you have follows chirality-Yes, Charges-Reduced, and EM-No, and goes to run the job on Gromacs software.

Hi all, I found a solution that worked for me: I did the energy minimization of my protein in vacuum first. Then, I proceeded with the solvation, ion addition and minimization without getting this issue. Hope it can be of help!

Dear Xabier

could you please share the mdp file which you used for vacuum minimization?

Fatemeh Haghighi

Hi Fatemeh, as far as I remember the parameters of the minimization were exactly the same. The only different thing was that the protein was not solvated!

I removed the periodicity of my .gro file with gmx trjconv and it worked.