How do some isocortex regions have such large ipsilateral connectivity?

[mruge]

I have been trying to create a computational model of the mouse isocortex using the connectivity matrix (Supplementary Table 3 from “A mesoscale connectome of the mouse brain” by Oh, S., Harris, J., Ng, L. et al.). I am focusing on only a few areas but when plotting the matrix for these areas there are some huge outliers. The paper mentions that there are only a few strongly connected areas but the difference is huge that I am doubting if what I am doing is correct, or some kind of transform might be needed. Especially the FRP → MOp and FRP → MOs connections are super strong in comparison to the others. Any help would be appreciated letting me know whether this should look how it does, or if I am doing something wrong and how to fix it

Thank you in advance!

[jeremyinseattle]

Hi @mruge,

I contacted one of the modeling experts from this study and here was his reply:

I would recommend using newer models which go beyond the purely linear model from Oh et al 2014.

1. Nadaraya-Watson based interpolation (more robust numerical method) model from: https://direct.mit.edu/netn/article/3/1/217/2194/High-resolution-data-driven-model-of-the-mouse
2. Nadaraya-Watson based interpolation model which includes information from cre lines, and uses significantly more data than the 2014 study: https://direct.mit.edu/netn/article/7/4/1497/117562

specifically, I would recommend using

mouse_connectivity_models/paper/connectivities/revision_thresholded/el_leafsurf_leafsmth_leafleaf_C57BL6J_080222.csv at 2020 · AllenInstitute/mouse_connectivity_models · GitHub

Please reply here if you have any follow-up comments or questions.

Best,
Jeremy