Komeil Jalali, may I ask what was your box size before and after NPT run? Also, what was your temperature?

Ali Ghamartale , I have run several simulations, and The result was either having an exploded system or having a system that has expanded in Z direction (lets say more that 10 times), and The temperature was 300 K

It seems that your systems contains strongly unfavorable potentials. You should visualize the trajectory, check the parameters and check which part is unstable.

Komeil Jalali So, I think the problem with density originated from your molecule's structure or maybe the number and the way you insert your molecules because in my view it is not common for the box dimension to change 10 times. You can check the exact dimension of box size at the end of .gro file.

could you please let me know how you build your molecule and making your topology file? did you use gauss view? did you optimize your molecule structure after building molecule? did you use gaussian for that? did you use online website for making topology or use the prepared could like mktop? how you make sure that the atom charges are calculated correctly?

Ali Ghamartale I tried to make my molecule in gussian view or even downloaded from the internet but unfortunately I encountered the error of "residue is not in the gromacs topology database", so I found an gro file for 250 toluene and choose the first toluene molecule for my initial molecule, yes you are right about the molecule structure but since I am using gromacs on a cluster I do not have access to the gromacs directory, so I wanted to know if you can suggest me any solution for that.

Komeil Jalali So about the structure, I suggest you to look at the following link: https://github.com/mosdef-hub/foyer/tree/master/foyer/opls_validation/toluene

about the gromacs directory on the cluster, if you are using MobaXterm and Graham cluster, just copy the following address in your directory section:

/cvmfs/soft.computecanada.ca/easybuild/software/2017/avx512/MPI/gcc7.3/openmpi3.1/gromacs

By the way, may I know what forcefield you are going to use in this project? Also, you did not mention how you made .itp and .top file. were they prepared or you made them? if you made them, how?

Feel free to ask me more question

Ali Ghamartale , I am grateful for your time and consideration, I am using Ubuntu cluster, I want to use gromos96 for my simulation, and about the topology file I downloaded it from the internet, and it was okay. but I think the problem is my Toluene gro file.

tol

15

1TOL C1 1 0.270 2.185 0.303

1TOL C2 2 0.138 2.224 0.256

1TOL C3 3 0.058 2.121 0.198

1TOL C4 4 0.097 1.994 0.199

1TOL C5 5 0.224 1.963 0.240

1TOL C6 6 0.311 2.052 0.295

1TOL H1 7 0.341 2.255 0.346

1TOL H2 8 -0.044 2.151 0.174

1TOL H3 9 0.028 1.913 0.179

1TOL H4 10 0.252 1.858 0.230

1TOL H5 11 0.401 2.011 0.339

1TOL C7 12 0.098 2.368 0.280

1TOL H6 13 0.108 2.405 0.382

1TOL H7 14 0.000 2.390 0.235

1TOL H8 15 0.157 2.434 0.215

0.0001 0.001 0.001

as you can see my problem is the residue type, so and I cannot run pdb2gmx for force field implementation.

Komeil Jalali Unfortunately, I can not understand what is wrong with the residue type but you can download toluene PDB file from following link:

https://github.com/mosdef-hub/foyer/tree/master/foyer/opls_validation/toluene

If you have gro file and topology file you do not need to run pdb2gmx. Just make sure charges and everything are correct in topology file, skip pdb2gmx step and go for making your box (gmx editconf or packmol software).

Ali Ghamartale thank you so much I will do as you told me to, I hope it work for me, I will tell you on the results of my work.

Sincerely

Komeil

No worries. I am interested to know about the result.

Ali Ghamartale , I Removed define = -DPOSRES from my npt file, and then everything was okay in my npt run, now I am waiting for the result of my md run. I will share the final result with you ASAP.

Sincerely

Komeil

Happy to hear that. Just a question, why did you put that in the first place?

Because before this I had a successful pdb2gmx and therefore I had to include this position restrain from the posre file, but in this case I couldn't use pdb2gmx and totally forgot to revise my npt file. And about the result of my simulation, unfortunately, IFT measured in my simulation was less than my previous md run.

Thank you so much for letting me know about your simulation. hopefully it works out

Ali Ghamartale

The force field that I intend to work with is GROMOS96 53A6 but I tried the toluene.gro and .top file that you obliged me and shared with me but I could not have a reasonable result and about the .gro and .top file that I had the result was not that much accurate that I wanted it to be. after that, I used .itp and .top files in the atb sever for the gromos force field (the united atoms), and the result was higher and now I am trying to modify the available files and if I do not have the desired results, I will try to use CHARMM in the next step, I wanted to keep you updated on my progress for now.

appreciate it if you share anything I can do to improve my work

Sincerely

Komeil

Komeil Jalali

Thank you so much for updating me. By the way, the .itp file that I told you was built for OPLS-AA forcefield. Although I have seen many studies with other forcefields, I recommend you to consider OPLS-AA if you are working with hydrocarbon systems. Hope you can solve your problem. Keep in touch and stay safe.

Ali Ghamartale I am trying to do bulk simulations of pure solvent, and check density/diffusion coefficient to calibrate my topology, but after energy minimization, the em.gro file only shows 1_I0L for every line instead of showing different numbers, so I encounter a fatal error in md simulation. I appreciate it if you let me know what is the solution.

Komeil Jalali

Actually, it has never happened to me, but I think the main issue is with your .itp and .top file. So, if you can provide me these two files plus your .mdp file, I may be able to try that by myself and find the issue.

By the way, I believe you have a little mistake in your toluene structure! The methyl group attached to the benzene ring miss hydrogens. So you may need to correct that and build your box again.

Ali Ghamartale , thank you for your reply, I have used the united structure molecule from atb and have modified the electron charges, and in the topology file you can find CH3 defined.

and again thank you for your time.

Komeil Jalali

I tried that with the files that you sent and it went well and toluenes are numbered correctly. By the way, I am using Gromacs 2019

Ali Ghamartale

I appreciate your time, I believe it is better to install gromacs 2019, because I am using 2020 version, and every time I encounter a lot of errors compared to 2018 version, I will surely keep you updated.

faithfully

Komeil

Ali Ghamartale , thank you I installed 2019.6 version and it works properlly. I will share my progress as soon as possible.

stay safe

Komeil

Ali Ghamartale, I worked on gromos force field and I have had some satisfying results, and now I am trying to compare it with the oplsa force field as you advised it is the best option to use for hydrocarbons, I managed to construct the molecules and generate the topol files and have had very accurate results for density but when it comes to self-diffusion coeficeion I face a significante error in its accuracy (between 50%to100%). I appreciate it if you kindly share your thoughts on this problem.