search

natverse / malevnc

Support for working with Janelia FlyEM Male VNC dataset
https://natverse.org/malevnc/
GNU General Public License v3.0
1 stars 0 forks source link

Write match json files for nblast dash #23

Closed jefferis closed 1 year ago

jefferis commented 3 years ago

@lisa-marin would like to initialise the LR matching w existing DVID matches before people get to work. Might you be able to have a go at this @dokato? I guess it would just be a matter of writing a suitable json file. But there might be some wrinkles if the match groups contain neurons that do not belong to the same folder in manc_matching_results

dokato commented 3 years ago

Ok, that's new, so far we've been talking about sth exactly opposite, see my first go here: https://github.com/flyconnectome/2021manc/blob/main/R/annotations/lrjson2clio.R (pushing json annotations to DVID)

lisa-marin commented 3 years ago

To clarify, we have discussed for some weeks the possibility of fetching and displaying existing matches in the nblast-dash when users open it to start to work.

Personally, I would prefer to assign the matches in nblast-dash independently of the assignments given by using other methods (and at an earlier stage of proofreading) and then check where there are discrepancies. Dominik is currently working on code to make comparisons with previously existing matches when we are ready to import and to flag up (but not override) any that differ.

A compromise would be to have previously assigned matches available to check while working with nblast-dash, and I think that's a good idea, but I wouldn't want them to appear the same way as the matches we assign using the tool. Does that make sense?

jefferis commented 1 year ago

closing as this is now unlikely to be relevant.