Diameter data overwritten in trajectory.gsd

[Damienfg]

Hi,

I'm using HOOMD to simulate the behavior of large spheres (colloids) in a bath of smaller spheres (solvent). A DPD framework is used along with a Lennard-Jones potential. I specified a diameter of 1.2 for the colloids and 0.08 for the solvent particles.

The simulation runs correctly, and I can read the outcome (trajectory.gsd) in Ovito.

However, using Ovito is annoying because I have to specify the diameters of the two types of particles every time I read the GSD file with Ovito. If my understanding is correct, this is because Ovito reads radius, not diameter.

Before switching definitely to another viewer like ChimeraX, I started examining the content of my GSD file to see if I could change something to make Ovito read a radius.

I noticed that the diameters I specified in my HOOMD Python code are overwritten in the trajectory.gsd file. All the diameter data are set to 1.0 (not 1.2 and 0.08).

I must be doing something wrong, but I don't know what.

Here is the part of the code dedicated to the gsd outcome :

# 🔹 Configuration de la sortie GSD
gsd_writer = hoomd.write.GSD(
    filename="/media/damien/T7/Simulation/Colloids/Final/LJDPD/trajectory.gsd",
    trigger=hoomd.trigger.Periodic(2000),
    mode='wb'
)
sim.operations.writers.append(gsd_writer)

I’m not copying the entire code to keep this message clear, but I can if necessary.

Many thanks for your help.

Damien

[joaander]

Add:

gsd.write_diameter = True

to include the diameters in your output.

[Damienfg]

Thank you, Joshua.

Where exactly should I insert the command line? Like this?
``
gsd.write_diameter = True

Configuration of GSD output

gsd_writer = hoomd.write.GSD(
filename="/media/damien/T7/Simulation/Colloids/Final/LJDPD/DPDconsLJ/Lecture_GSD/trajectory.gsd",
trigger=hoomd.trigger.Periodic(100),
mode='wb'
)
sim.operations.writers.append(gsd_writer)
``

Also, do I need to include the following at the beginning of the code?

import gsd.hoomd

Finally, my code does assign the diameters of both the colloids and the depletants:

snapshot.particles.diameter[:N_coll] = D_coll # Colloids
snapshot.particles.diameter[N_coll:] = 0.08 # Depletants

However, regardless of the values I assign, another script I use to read the trajectory.gsd file always outputs:
particle diameter: [1. 1. 1. ...]

I'm not sure what I'm doing wrong — the simulation itself behaves as expected, which makes this even more confusing. I just can't seem to get the correct diameters written to the GSD file.

Thanks again for your help (and for HooMD, which is truly a fantastic tool!).

[joaander]

You can place gsd_writer.write_diameter = True any time after you construct the gsd_writer object. The attribute is documented here: https://hoomd-blue.readthedocs.io/en/v5.1.1/hoomd/write/gsd.html#hoomd.write.GSD.write_diameter

You do not need to import gsd.hoomd.

[Damienfg]

Thank you so much, Joshua! I followed your approach and it worked perfectly. I also fixed an issue in the GSD reading script that was causing problems. Everything is running smoothly now. My next step is figuring out how to display particles in Ovito with the specific diameters I defined in my simulation.

[Damienfg]

Everything appears to be working correctly in Ovito now. The diameter being written in the GSD trajectory file, the particles are properly sized in the Ovito viewer.
Thanks again for your help on this.
Damien