Usability needs for MuPT release?

[mrshirts]

Hi, all - use this issue thread to discuss ways that you want new users to be able to interact with MuPT functionality. What scripts and tools need to exist to make it easier for people to use the functionality? What problems have you had? What things do you think we should have explanations/example for? We'll use this feedback to create some todos for the (soon to be created) usability subgroup.

[joelaforet]

I think we should have a ReadTheDocs page for new users.

[joelaforet]

I also propose we have an "install-benchmark", where one person from each university (so we get different HPC environments + machines), goes through the process of installing MuPT and makes note of any friction they encounter. They should record how long it takes to go from "git clone mupt" to getting a working Python instance they can use on their cluster/ Python environment.

[sbkashif]

What scripts and tools need to exist to make it easier for people to use the functionality?

Entry point should be minimal to invite a wider user base

I would advocate for a centralized MuPT config file (assuming cli run, similar adoption in python notebooks) to define all the parameters needed for a run, while keeping the required entries minimal (and resonable defaults for the rest). The minimal entries needed to run a basic polymer build and analyze it could be:

• monomer smiles,
• reaction template (rdkit for now, mupt repr later),
• processing conditions (limit to T, assuming NVT ensemble to be default),
• resolution (aa or cg),
• molecular simulation engine,
• force field,
• termination criteria (target chain lengths, degree of cross-linking, or any function available with analysis toolkit).

Ideally, we can have blocks in the config file that correspond to different sub-tools, like builders, reactions, and analysis, and have entries from the central config file routed to their individual config files. One exception could be the analysis config file (which can have an independent entry point) since some users would just want to selectively use our analysis tool on polymers they built elsewhere.

Easy to customize

Once users get a hang of the tool, they can be pointed to readthedocs for more options, such as changing the force field, ensemble, etc. This part would help user select from available options to customize their workflow.