flow comparison - Githubissues

neurodata / maggot_models

Modeling Drosophila larval connectomes

BSD 3-Clause "New" or "Revised" License

5 stars 6 forks source link

flow comparison #91

Open bdpedigo opened 3 years ago

bdpedigo commented 3 years ago

[ ] try to get significance of feedforwardness somehow
- jovo said just mass above 0 in my diagonal plot thingy
[ ] compare what you get from signal flow to WalkSort
[x] compare each of the adjacency matrices (sorted) to each other when sorted by each of the other sortings...
- [x] 5 x 5 diagonal weights
- [x] 5 x 5 of adjacencies

bdpedigo commented 3 years ago

[ ] want to test which are feedforward and how much
- like the result about A -> A is not feedforward
- question - but this is always relative to A -> D?
[ ] need to think of a better null model for the joint graphs

bdpedigo commented 3 years ago

[ ] what happens if we remove homotopic connections esp in the A -> A

bdpedigo commented 3 years ago

[x] try making the second row of that figure using WalkSort

bdpedigo commented 3 years ago

[ ] modify WalkSort to be front-back-walk-sort

bdpedigo commented 3 years ago

Write up WalkSort 1 - backwards walk sort? On the right it looks really diffuse Fraction of the mass above 0 as a metric Could fit some kind of feedforward model… Simulated data plot with walk sort

[ ] Ratio of A-D in upper triu / A to A in upper triu as a function of different sortings