Dev Meeting 230517

[ehhartmann]

Goal : Create project plan for releasing KIMMDY v2.0

Requirements stated in 221212 Meeting:

• Using KIMMDY/Release
  - Partial charges should be good
  - Espaloma parameters as new verions
  - Publishing: - Methods, - one use case / example like hexala for statistics and details , -one big one to show off, - maybe use ala-ala-dopa-ala-ala
  - Documentation: - example, - auto generated technical documentation
  - Features: ,- Espaloma -> next paper ,- Collagen -> next paper ,- MC w/ rejection + testing -> first release!

Tasks

features:

• finish KMC implementations Eric, Kai
• calculate all necessary partial charges Eric
  • still missing but not necessary for now: terminal charges, crosslinks, some His species (maybe combine with extending radical QM dataset)
• get partial charges into KIMMDY (validate) Jannik
• finish relaxation schemes (slow growth and hydrogen placement Eric
• implement/ improve choice of starting structure for relaxation Eric, hat predicted
• clean up config file (and config.py) Jannik, Kai
• write analysis tools for kimmdy all
  • improve trajectory concatenating tool to get meaningful trajectories based on KMC scheme
  • graph/vmd visualisation of population of radical positions in the molecule
  • graph/vmd visualisation of radical flux from each position
  • graph/vmd visualisation of most probable radical migration pathways
  • analysis of racemization
  • plot energy profile over simulation (of protein only)
• HAT reaction plugin Kai

testing

• fix tests
• extend tests
• build examples from tests?

go through issues

documentation:

• tidy up Jannik
• add docstrings
• clean up examples Kai

code review

publishing:

• do simulations of toy systems
• analyse simulations
• write

[KRiedmiller]

• Reaction API: whats the difference between ReactionOutcome and ReactionResult and should there be any?

[jmbuhr]

One ReactionResult contains a list of potential ReactionOutcomes (which is a rate and a recipe)

[jmbuhr]

naming things is hard 🙈

[KRiedmiller]

But ReactionOutcome is itself already a list of reactions recipes, rates and framenumbers. We're just adding an empty dimension (in numpy terms)
(also talking about the newest 'betterKMC branch')

[ehhartmann]

We have a little more than one week left before the meeting with Frauke. I would like to focus on some parts to have a working version until then (also, we could present kimmdy at the retreat).
If possible, having Leif's parameters in kimmdy would be great. Also, some basic visualization/analysis tools and a working HAT plugin would add a lot. Do you think its possible to finish these features until then @jmbuhr @KRiedmiller ? Would you add/remove something?

[jmbuhr]

I want to finish some cleaning-up today and add more tests in the process so that I see what breaks when I add the parameters, but if this goes well we can move pretty fast.
I also added markdown support for the docs and an example of a quarto document that get's rendered to markdown, to then be picked up by the docs, so you can add visualizations to this easily, though this is still a bit flimsy and lately sphinx is ignoring the folder with the image outputs... But if I get this to work properly we can show off analysis tools very nicely.