Unveiling and Steering Connectome Organization with Interpretable Latent Variables
This work provides a novel tool for neuroscientists to analyze brain architecture and potentially design bio-inspired artificial neural networks, though it is incremental in applying existing representation learning techniques to connectomics.
The authors tackled the problem of understanding the brain's complex connectome by developing a framework that extracts interpretable low-dimensional representations from the Drosophila connectome, achieving effective graph reconstruction and controllable generation of subgraphs with predefined properties.
The brain's intricate connectome, a blueprint for its function, presents immense complexity, yet it arises from a compact genetic code, hinting at underlying low-dimensional organizational principles. This work bridges connectomics and representation learning to uncover these principles. We propose a framework that combines subgraph extraction from the Drosophila connectome, FlyWire, with a generative model to derive interpretable low-dimensional representations of neural circuitry. Crucially, an explainability module links these latent dimensions to specific structural features, offering insights into their functional relevance. We validate our approach by demonstrating effective graph reconstruction and, significantly, the ability to manipulate these latent codes to controllably generate connectome subgraphs with predefined properties. This research offers a novel tool for understanding brain architecture and a potential avenue for designing bio-inspired artificial neural networks.