search

open-physiology / apinatomy-models

Working and published ApiNATOMY models
2 stars 4 forks source link

Connectivity of Ardell-Armour neuron 6 path doesn't match model #25

Open dbrnz opened 1 year ago

dbrnz commented 1 year ago

Path 6 is shown as having branches originating in both the left and right atria and passing via the atrial ICG towards the spinal cord, and other branches originating in both the left and right ventricles and passing via the ventricular ICG towards the spinal cord. However connectivity obtained from SCKAN for the path:

  1. does not show branches passing through the cardiac ICGs.
  2. does not have the left atrium (UBERON:0002079) in the result set obtained from the neru-6 query.
Screen Shot 2022-09-17 at 12 25 34 PM
Screen Shot 2022-09-17 at 12 24 46 PM
tgbugs commented 1 year ago

Point 1 is expected and is a matter of the overspecification of aacar-6 as discussed in https://github.com/open-physiology/apinatomy-models/commit/19d6f32a5fdf278ac3e81261516a8f355deae007. The drawing in the diagram is consistent with multiple different possible paths, the one we represent is the simplest one that is consistent with all connections and axon locations.

Will investigate Point 2.

tgbugs commented 1 year ago

For point 2 it seems that that chain is missing from the source sheet so will add it.

tgbugs commented 1 year ago

the problem was the for ac-sn-la2_6 an-sn-la1_6 the identifier was A76 which was redundant for right ventricle, it should have been and now is A77 which is left atrium, the duplicate structure meant that when we did our count check we did not catch the issue because the number of processes was correct