New Barcoded Rabies Technique Maps Hundreds of Neural Connections at Once

Understanding how the brain works requires knowing which neurons talk to which other neurons—the wiring diagram of thought itself. For decades, neuroscientists have mapped these connections one painstaking cell at a time, a process so labor-intensive that researchers could typically trace only the inputs feeding into a handful of nearby neurons before exhaustion or funding ran out. A team working at the Francis Crick Institute has now developed a technique that shatters this bottleneck, allowing them to map the incoming connections of hundreds of neurons simultaneously while keeping track of where everything sits in the tissue.

The method, called Barcoded Rabies In Situ Connectomics or BRISC, weaponizes rabies virus in a clever way. The researchers engineered rabies viruses to carry random molecular barcodes—think of them as genetic ID tags—and then used these tagged viruses to infect neurons in the mouse brain. When a rabies virus jumps from one neuron to a connected neuron, it carries its barcode along. By reading which barcodes appear in which cells and matching them between starter neurons and their upstream partners, the researchers can reconstruct the circuit. The key innovation was creating viral libraries diverse enough to label over 1,000 neurons with distinct barcodes, then controlling how densely those starter neurons were packed to prevent barcode mixing that would muddy the results.

They tested BRISC in the primary visual cortex of mice, the brain region that first processes what the eye sees. The results were striking: they mapped the inputs of 385 neurons and identified 7,814 putative synaptic connections—the physical junctions where one neuron sends a signal to another. Because they read out the barcodes while the tissue remained intact, they preserved the spatial architecture of the circuit, seeing not just which neurons connect but where those connections originate in the layered structure of the cortex. The data revealed patterns: neurons in certain layers preferentially received inputs from specific cell types, and long-range connections from distant brain regions showed a topographic organization, meaning neurons that are close together in space tend to receive inputs from nearby locations in their source regions.

What makes this work significant is the scale and speed. Previous connectomics methods, like electron microscopy reconstruction, can map circuits with exquisite detail but require months or years of manual tracing for a small volume of tissue. BRISC trades some ultrastructural detail for the ability to see hundreds of neurons at once, revealing the organizational logic of circuits rather than just their anatomy. The technique preserves spatial information—a critical advantage over purely sequencing-based approaches that lose the where—while gaining the throughput that sequencing provides.

The implications ripple outward. Visual cortex is just the beginning. The same approach could be applied to other brain regions and other animal models, potentially accelerating the construction of connectomes—complete wiring diagrams—across the nervous system. Neuroscientists have long suspected that circuit organization follows rules, that the brain is not random but structured according to principles that evolution has refined. BRISC offers a tool to test that hypothesis at scale, to map enough circuits fast enough to spot the patterns. It will not replace detailed anatomical studies, but it changes what becomes possible: systematic, high-throughput mapping of how neural circuits are built and how that architecture relates to computation.