chemicaljson
chemicaljson copied to clipboard
OpenChemistry
Example with multiple coordinate sets / animations?
Bump. There doesn't seem to be any way to support multiple coordinate sets in CJSON, which is a big problem for animations, dynamics, conformers, etc.
One good way to introduce a series of coordinate sets would be to introduce some kind of iterator, including a keyword in the geometry identifying the type of coordinates, xyz, unit cell, and if this is a coordinate series (could be both xyz or unit cell in case of ab initio molecular dynamics). We will need to add a time stamp (from MD) or an iterator integer to ensure we can extract them in the correct time order.
Bert
On Feb 12, 2017, at 6:16 PM, Geoff Hutchison [email protected] wroteBump. There doesn't seem to be any way to support multiple coordinate sets in CJSON, which is a big problem for animations, dynamics, conformers, etc. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.
Are we going to address this in CJSON or work on the new QC JSON and handle it there?
I was going to get this in here, but got distracted by switching to a more modern JSON library. I think what we do here/in QC JSON will hopefully share a lot. Thanks for the reminder, it should not have taken this long, apologies! I think this would offer a simple way of getting smaller trajectories out of other packages in that case too.
This seems to already be in the wild as part of 3dSets .. an array of arrays.
What's missing:
- time step (from MD)
- relative energies or other properties (e.g., recording forces through an optimization or MD)
- a way to indicate the default set (e.g., usually with geometry optimization, the default
3dset would indicate the best, lowest-energy geometry .. but in MD, the default might duplicate the first coordinates.)- In other words, if
3dSetsis available, can something else indicate the default index?
- In other words, if