Paper roller coaster

[Dario55]

I have listed below a high order proposal of what I think is missing for the paper and tentatively suggesting who I think could be responsible for each of the bits. Especially within Neurogears, of course you decide if this task allocation make sense and how to allocate tasks between you guys. This is just a proposal.

In the interest of not adding extra meetings for this, I would propose to turn task control into a paper meeting (basically to track progress on the first part below), and data architecture into a documentation meeting, so DJ will also be there in case we need their input. This just until the deadline. We should agree together on which one of the things below are actually doable and on realistic deadlines for them. I would suggest as a overall deadline to submit the preprint December 15th. I have divided the tasks in 3 main cathegories: PAPER, DOCUMENTATION, DATA

Please feel free to edit comment as appropiate

PAPER (deadline: 15.12.2024)

• Paper main text @Dario55 / @jkbhagatio / @glopesdev
• Paper methods @Dario55/ @jkbhagatio @glopesdev @anayapouget zimo
• Paper figures @Dario55 / @jkbhagatio
• implement comments from SC meeting and sfn @Dario55 (not going to write all of them here or will become a book)
• finish cleaning up @Dario55
• for 24/7 foraging figure check tracking from Tom/Andre and Zimo. @Dario55 If needed I would need help from @aspaNeuro @NeuroThom to retrack.
• Generate social figures. Replicate what’s done for poster with more mice plus few extra analysis discussed for cosyne poster (let’s see what we have and then we decide where to draw the line between this and social paper). @jkbhagatio
• look at foraging efficiency conditional on dominance @jkbhagatio
• in general we need to make sure of data integrity and quality: for at least 6 pairs (very minimum to do any statistic) we need to have high quality tracking and identity, tube test before and after social. @jkbhagatio
• make sure that effect we saw like similar patch occupancy or high coforaging rate are real and not due to if tracking artefacts. @jkbhagatio
• make sure that database has all the data we need @jkbhagatio
• rerun and check dominance based on corridors encounteres for all 6 pairs @jkbhagatio @anayapouget
• decide whether we want to also use fighting/fast chasing as a metric for dominance or not. @jkbhagatio @anayapouget
• fix rfid readings that in the poster where all over the places. Maybe this will sort itself when fixing the tracking @jkbhagatio
• Get quantification of model prediction accuracy for few mice (ideally the six pairs) to show that model work. At the moment we have only one example @jkbhagatio @anayapouget @lochhh

DOCUMENTATION

Svenja will present the full and detailed plan for documentation/website soon, Here I just want to give an overview so people can have an idea of how much involvement we are hoping to have from each team member. We would need help from everyone . I have tentatively put some names here and there of who I think knows the relevant information for each step, but please feel free to edit. @snierwetberg will be overseeing and working on all aspects of the docs/website, so I haven’t written her name every time.

• Mechatronic documentation @lorycalca .
• wiring diagram @lorycalca @aspaNeuro
• Implementation of modular bonsai: @glopesdev ,@aspaNeuro
• module workflows/How to/glossary bonsai: @NeuroThom @glopesdev ,@aspaNeuro
• python API for modules, how to, glossary: @NeuroThom @glopesdev, @jkbhagatio ,@lochhh
• Analysis for modules, how to, glossary from API @NeuroThom @glopesdev, @jkbhagatio ,@lochhh @anayapouget
• Analysis for modules, how to, glossary from DJ, @jkbhagatio ,@lochhh @anayapouget, DJ
• How to experiments:
  • How to start an experiment: all the steps required to run an experiment front to end (check list). @lorycalca , @jkbhagatio , @aspaNeuro

DATA

If we are going to share data, we need to:

• document the data structure. This can be done in the form of a DJ database, raw data or both.
• document the experiments done, the mice, etc. We can use one of the social for this to start with, and then go back to previous experiments. @jkbhagatio, @glopesdev.
• if we are sharing a big proportion of the data we need to ensure data quality, compliance with the rules if we show videos.@jkbhagatio, @glopesdev,@Dario55, @lorycalca
• find where to store them @jkbhagatio, @glopesdev