Energy minimization of a non orthorhombic surface

User discussions
1

GROMACS version: 2021.4
GROMACS modification: Yes/No
Hello everyone,
I am trying to minimize the energy of a system comprising of a non orthorhombic surface with an ionic liquid on top of it. However, the results after the minimization is wrong. Can anyone help me to understand how to address this issue? Thanks in advance.
Arka Prava Sarkar
Ph.D. Scholar
IIT Kharagpur

2

Hi, I wonder what the specific error of the result is? If it is the error of task running, gmx would post the relevant information in the last part of .log file.

3

Hi Seyilaxa,

Thanks for your response. There is no runtime error, but when I visualise the gro file after minimization, it turns out that the surface becomes more like an orthorhombic one, but the box vectors have wrong values. Thanks.

Thanks & Regards,
অর্ক প্রভ সরকার
Arka Prava Sarkar
বৃত্তিজীবী বাচম্পত্যাধয়নকারী
Research Scholar (Ph.D.),
গণনা এবং তথ্য বিজ্ঞান কেন্দ্র
Centre for Computational and Data Science (CCDS)
ভারতীয় প্রযুক্তি প্রতিষ্ঠান, খড়গপুর
Indian Institute of Technology, Kharagpur
পশ্চিমবঙ্গ-721302
West Bengal -721302
India
Email1 : sarkar.arkaprava200@gmail.com
Email2 : arkaps.ccds@iitkgp.ac.in

4

There are many possibilities that could lead to this. I’m sorry I could not give a determined solution because I don’t know the detail of your system. I suggest you check the parameters of simulation which describes the interaction between particles, and confirm if PBC conditions are suitable.

5

I am attaching the input files and the output gro file with this email.

Thanks & Regards,
অর্ক প্রভ সরকার
Arka Prava Sarkar
বৃত্তিজীবী বাচম্পত্যাধয়নকারী
Research Scholar (Ph.D.),
গণনা এবং তথ্য বিজ্ঞান কেন্দ্র
Centre for Computational and Data Science (CCDS)
ভারতীয় প্রযুক্তি প্রতিষ্ঠান, খড়গপুর
Indian Institute of Technology, Kharagpur
পশ্চিমবঙ্গ-721302
West Bengal -721302
India
Email1 : sarkar.arkaprava200@gmail.com
Email2 : arkaps.ccds@iitkgp.ac.in

field.top (20.6 KB)

(Attachment config.pdb is missing)

(Attachment em.gro is missing)

em.mdp (1.23 KB)

6

Dear APSDft,

The .gro file is missing but I guess you want to build a non-orthorhombic surface by Au and Pt which are frozen to keep the surface. In your decription the freeze does not work, resulting in the surface deformation. There is a similar topic in the forum which might help you (https://gromacs.bioexcel.eu/t/help-with-freeze-groups-in-gromacs-holding-a-surface-fixed-in-space/4934/2)

Therefore, using gmx make_ndx to make a .ndx file including the all Au and Pt and use this .ndx to specify the freeze group with energy group exclusions maybe could solve the problem. Additionally, if this also doesn’t work, I think use position_restraints in .top file to fix atoms would also be a resonable method.

7

Hi,
I tried according to your suggestions, but the error still persists. Since gromacs community is not supporting uploading the pdb and gro files, I am sending you the corresponding input and output structure xyz files

(Attachment out.xyz is missing)

(Attachment in.xyz is missing)

8

Hi,
Gromacs also rejected xyz files. Please find the dat extension files. You can just change the extension to xyz and visualise in vmd

out.dat (18.2 KB)

in.dat (18.2 KB)

9

I have read your structure file, I found that you used a dodecahedron box as the PBC condition and the atoms that make you doubt are at the boundary of the box. The only thing you need to do is use command pbc box -on in VMD to check the atom in .gro is still at the boundary. An atom at the boundary of box is equivalent to being on the opposite side of box, so actually the Au atoms have not moved, and this would not be an issue in the next step simulation (but if you keep using freeze you would not be able to use thermostat in NPT simulation).

10

Hi,
I want to do an REMD simulation with the energy minimized structure to sample different conformers of the ionic liquid on the surface. I think I am safe to go ahead with this then?

Thanks & Regards,
অর্ক প্রভ সরকার
Arka Prava Sarkar
বৃত্তিজীবী বাচম্পত্যাধয়নকারী
Research Scholar (Ph.D.),
গণনা এবং তথ্য বিজ্ঞান কেন্দ্র
Centre for Computational and Data Science (CCDS)
ভারতীয় প্রযুক্তি প্রতিষ্ঠান, খড়গপুর
Indian Institute of Technology, Kharagpur
পশ্চিমবঙ্গ-721302
West Bengal -721302
India
Email1 : sarkar.arkaprava200@gmail.com
Email2 : arkaps.ccds@iitkgp.ac.in

11

I actually used a triclinic box using gmx editconf, then after the topol.gro file is generated from mdrun, I used the trjconv to convert it back to triclinic box.

Thanks & Regards,
অর্ক প্রভ সরকার
Arka Prava Sarkar
বৃত্তিজীবী বাচম্পত্যাধয়নকারী
Research Scholar (Ph.D.),
গণনা এবং তথ্য বিজ্ঞান কেন্দ্র
Centre for Computational and Data Science (CCDS)
ভারতীয় প্রযুক্তি প্রতিষ্ঠান, খড়গপুর
Indian Institute of Technology, Kharagpur
পশ্চিমবঙ্গ-721302
West Bengal -721302
India
Email1 : sarkar.arkaprava200@gmail.com
Email2 : arkaps.ccds@iitkgp.ac.in

12

Using trjconv would not influence the PBC condition, so just go to the next step.

13

Thanks for all the help and discussion. I will proceed to the next step. many thanks.