Edis Jakupovic

I’ll be home in ~1.5 hrs, can take a look then

That seems like the best short term fix to me. There was a bug fix listed here https://docs.h5py.org/en/latest/whatsnew/3.4.html relating to the `core` and `stdio` drivers and using `mode='a'`, however `H5MDWriter`...

@orbeckst can I ask you about the multiple groups in `/particles/` - is there any case that makes sense for a `mda.Universe` to load multiple trajectories from multiple particle groups...

@richardjgowers @lilyminium thanks for the replies. I do have a `_format_hint` defined [here](https://github.com/MDAnalysis/mdanalysis/blob/840a977294311555232114fe998163ce53a88716/package/MDAnalysis/coordinates/H5MD.py#L504-L509) in H5MDReader. So is the issue that the `h5py.File` is being picked up by the ChainReader format...

yes, this is related to point 1) in #4561 basically, `/particles` can contain arbitrary "atom groups" like `/particles/group_1/`, `/particles/group_2/`, etc... that contain the position, velocity, and force data for specific...

the same thing is happening in #4320

> and note that @PythonFZ 's H5MD file cannot be loaded by pyh5md, presumably because it is missing the **mandatory** h5md root level group (see https://www.nongnu.org/h5md/h5md.html#h5md-root-level ). that's interesting... should...

Hi @RMeli, thanks for the clarification! I ended up just adding them in my conclude method like this and it works exactly how I wanted ```python def _conclude(self): self.results.frames =...