Documentation of mouse_connectivity_models 0.0.1


This package is currently in its early Beta stages and is under active development.

mouse_connectivity_models (mcmodels) is a Python package providing mesoscale connectivity models for mouse using data from anterograde viral tracing experiments. This package was initially written when developing the latest voxel-scale connectivity model for the publication ‘High resolution data-driven model of the mouse connectome’, and has since become a repository for the various connectivity models we are researching.

Information for Researchers interested in the latest voxel-scale model

The latest voxel-scale model is a weighted, directed graph represented as an adjacency matrix in which:

  • rows : set of source voxels at 100 micron
  • columns : set of target voxels at 100 micron
  • elements : the nonnegative connection weight from a given source voxel to a given target voxel

This matrix has on the order of ~ 200,000 by 400,000 elements and most likely will not in memory. However, our model learns a low-rank representation of this adjacency matrix that we utilize to construct the full matrix in pieces on demand through the VoxelConnectivityArray class. For more information see the Data wrangling section of the User Guide.

To access the voxel model without refitting, visit this download page.

Regionalized voxel-scale model

Additionally, one may be interested in analysis of the model at a regional resolution. We have available a set of adjacency matrices of the connectivity between a set of 293 summary structures for which we integrated our voxel-scale connectivity. These can be conveniently downloaded (and subsequently loaded from your local machine) through the VoxelModelCache class.

Download the regional models here (CSV files).

Level of Support

We are not currently supporting this code, but simply releasing it to the community AS IS but are not able to provide any guarantees of support. The community is welcome to submit issues, but you should not expect an active response.



  • Joseph Knox
  • Kameron Decker Harris
  • Jennifer Whitesell
  • Nile Graddis


None yet. Why not be the first?