Couldn't load bonded topology of water molecules ?

[Kamecs]

Dear MDAnalysis Developers.

Thank you for your providing of useful MD trajectory analysis tool.
This tool has greatly accelerated my trajectory analysis.

I am experimenting with using ProLIF (https://github.com/chemosim-lab/ProLIF), which references MDAnalysis.
Recently, ProLIF has enabled water bridge analysis, so I created a universe object as a system that includes water and passed it to ProLIF.
However, following error message was shown:
rdkit.Chem.rdchem.AtomValenceException: Explicit valence for atom # 1 H, 2, is greater than permitted

Since no error occurs without water, it appears that the molecular topology is ignored and the system determines that the hydrogen in water is bonded to a nearby oxygen.

I have attached the original topology and coordinate file without the error, so please check whether the PSF topology has been properly imported into MDAnalysis.

Sorry for the trouble.
Best regards.

to_mda.zip

[orbeckst]

Hello, this is not an issue but a question. I'll transfer your question to the discussion forum.

[orbeckst]

If you're new to MDAnalysis, read the User Guide. This will also help you when working with other tools (such as ProLIF) that use MDAnalysis.

To check that your files can be read, load the, into a Universe, like

import MDAnalysis as mda
u = mda.Universe(topology_file, trajectory_file)

print("Universe atom group", u.atoms)
print("Trajectory", u.trajectory)

If this works then your data can be read and further problems are probably best discussed in the ProLIF discussion forums.

[Kamecs]

Sorry for the late reply.

I've been using MDAnalysis for several years.
We have not analyze water molecules seriously, and this is the first time that we are tackling water molecule analysis with prolif.
However, the problem on the prolif matter has been resolved after trying various things behind the scenes.

Sorry for the inconvenience.
Thank you for your kind response.

Best regard.