Re: [h5md-user] topology

[Pierre de Buyl]

Hi Felix,

On Wed, Apr 23, 2014 at 10:00:17AM +0200, Felix Höfling wrote:
> Am 22.04.2014, 13:59 Uhr, schrieb Pierre de Buyl
> <address@hidden>:
> 
> >I show below a "h5ls -r" dump of a topology group:
> >"""
> >/chains                  Dataset {4000, 2}
> >/crosslinks              Group
> >/crosslinks/step         Dataset {100/Inf}
> >/crosslinks/time         Dataset {100/Inf}
> >/crosslinks/value        Dataset {100/Inf, 417/Inf, 2}
> >"""
> >
> >Here, "chains" is a "fixed in time" element of 4000 pairs and
> >"crosslinks" is a
> >"time dependent" element with 100 time steps and at most 417 pairs.
> >
> >The position in "/topology/<group>" could be (in my opinion) changed to
> >"/particles/<group>/topology". I wanted to work with minimal
> >modifications,
> >which why I use "/topology" now but putting the information in
> >"/particles"
> >seems more natural to me.
> >
> Indeed, the mailing list (and
> http://nongnu.org/h5md/discussion.html) is a good place to share
> ideas on extensions of H5MD.
> 
> Some time ago, Olaf told me about his thoughts on how to store
> topological information. I think this is pretty much in line with
> your proposal. Olaf's ideas were pretty much inspired by what is
> stored in the PDB format.

The pairs correspond to the "CONECT" lines in PDB if I understand correctly. I
don't know whether triples and quadruples are considered.

> What I'm missing in your post is what information is actually
> stored? Am I right that you define a pair bond by a pair of particle
> IDs? In /particle these IDs are unique only within a subgroup so
> bonds across these subgroups would be difficult to implement. (They
> may simply not be supported, requiring the subgroups to be
> sufficiently large. E.g., they would need to hold the complete
> macromolecule not only parts of.)

I store pairs (hence the '2' dimension in the datasets) of particles indices.
This has to occur per group as there is no uniqueness of indices among groups,
as you noticed. I did not take into account the situation where there are bonded
interactions among separate groups. Would that be relevant to anyone?

Also, what I consider is that the particle id refers to the index of the
particle in the '/particles' datasets if "id" is absent and to the values of
"id" if it present.

> We better keep the topology separate from the /particles tree. The
> information complements the information in /particles, but is quite
> different in structure. In /particles, there is one entry per
> particle while the topology is about subsets of pairs, triples, etc.

It is true that the information in /topology is structured quite differently
than in /particles. I don't see a hard reason to choose any of the solutions
yet, though. An advantage of putting it in /particles could be that the
particles' positions and topologies could be copied more simply from a file to
another.

I realize that I forgot also that topology information could be pairs of
particles combined with an equilibrium/constraint distance. I have no idea on
how to add that yet.

Cheers,

Pierre