Computational model of developmental self-assembly of a cortical microcircuit
In the authors' words
Cortical microcircuits emerge from local developmental interactions, yet large-scale computational models typically prescribe connectivity rather than explain its formation. Here we use BioDynaMo to simulate the self-assembly of ~70,000 cortical neurons from 1,200 progenitors in 3D physical space through migration, neurite growth and contact-based synapse formation. With one calibrated filling fraction per connection type, the locally formed synapses reproduce experimentally measured connectivity across 64 connection types (mean absolute error 0.0062) without prescribing individual connections. The grown spiking networks sustain stable, inhibition-dominated activity, and the same developmental rules, given different targets, generate both a canonical cortical microcircuit and a visual-cortex variant. Inverting the migration programme produces a reeler-like column whose aberrant connectivity destabilises network activity. Together, these findings provide a mechanistic link between local developmental processes and the emergence of cortical anatomy, connectivity and function.
In plain words AI summary
The authors used BioDynaMo to simulate how about 70,000 cortical neurons self-assemble from 1,200 starting cells in 3D space. The simulated neurons formed synapses through migration, neurite growth, and contact-based synapse formation. With one calibrated setting per connection type, the model reproduced experimentally measured connectivity across 64 connection types with mean absolute error 0.0062. The grown networks showed stable, inhibition-dominated activity, and the same rules produced both a standard cortical microcircuit and a visual-cortex variant. Reversing the migration program created a reeler-like column with abnormal connectivity that destabilized network activity.
Written by Kimi K2.5 using only the abstract above. Jev checked that every sentence is supported by the abstract sentences it cites, and removed any that were not. Point at a sentence to highlight its source.
Appeared: Wednesday, November 11. bioRxiv. Preprint, not yet peer-reviewed.