Connectome-seq Turns Brain Wiring into a Sequencing Problem, Mapping Thousands of Synaptic Connections with RNA Barcodes
University of Illinois researchers publish Connectome-seq in Nature Methods, using RNA barcodes to map over 1,000 neurons at single-synapse resolution and uncover previously unknown connectivity patterns in the mouse brain.
Overview
A team led by Boxuan Zhao at the University of Illinois Urbana-Champaign has developed a technique that converts the problem of mapping neural connections into a high-throughput sequencing task. The method, called Connectome-seq, uses RNA molecular barcodes to identify which neurons are physically wired together at their synapses, according to a paper published in Nature Methods. In a proof-of-concept experiment, the researchers mapped more than 1,000 neurons in a mouse brain circuit and discovered connectivity patterns that had not been observed with existing methods.
What We Know
Connectome-seq works by assigning each neuron a unique RNA barcode. Engineered proteins then carry those barcodes from the neuron’s cell body to the synapse, the junction where one neuron communicates with another, and anchor them there. Researchers isolate the synaptic junctions and use high-throughput sequencing to read out which pairs of barcodes ended up together, revealing which neurons are directly connected, according to the University of Illinois.
Zhao described the process with an analogy: imagine balloons covered in barcode stickers, with some stickers traveling down the strings that tie two balloons together. “If two balloons are tied together at the end, the two barcodes meet at the junction,” Zhao said. “Then we snip out the knots and sequence the barcodes,” according to the University of Illinois.
The team validated Connectome-seq in the mouse pontocerebellar circuit, a pathway that connects two distinct brain regions. The mapping revealed previously unknown connectivity patterns, including direct connections between cell types that were not known to be wired together in the adult brain, according to ScienceDaily. Through integrated analysis of connectivity data and gene expression, the researchers also identified molecular markers that are enriched in connected neurons, according to the Nature Methods paper.
The key advantage over existing approaches is scale and speed. Traditional connectomics relies on electron microscopy, which requires physically slicing and imaging thin sections of brain tissue, a process so labor-intensive that fully mapping even a small volume can take years. Newer sequencing-based methods can label many neurons but typically trace only projections rather than specific synaptic partnerships. Connectome-seq bridges this gap by delivering single-synapse resolution at throughputs that existing technologies cannot match, according to the University of Illinois.
“This enables simultaneous mapping of thousands of neural connections with single-synapse resolution, a capability that doesn’t exist in any current technology,” Zhao said, according to the University of Illinois.
What We Don’t Know
The current validation was performed in a single mouse brain circuit. Whether Connectome-seq can scale to map connectivity across entire brain regions or across species with larger, more complex brains remains to be demonstrated. The technique relies on adeno-associated virus delivery and engineered proteins, and any limitations in how uniformly these tools label neurons across different brain areas could introduce blind spots.
It is also unclear how the method performs in diseased tissue, where synaptic architecture may be disrupted. The researchers have proposed that Connectome-seq could enable comparisons between healthy and diseased brains, potentially identifying circuit changes before symptoms appear in conditions like Alzheimer’s disease, according to ScienceDaily. But that application has not yet been tested.
Analysis
Connectome-seq addresses a genuine bottleneck in neuroscience. The field has long understood that brain function emerges from patterns of connectivity, but mapping those patterns at synaptic resolution has been prohibitively slow. Electron microscopy projects like the Allen Institute’s mouse brain connectome and the FlyWire project for Drosophila have demonstrated what is possible, but each required years of effort and massive computational resources for relatively small volumes of tissue.
By reframing connectivity as a sequencing problem, Connectome-seq taps into the same economies of scale that have driven down the cost of genomics over the past two decades. If the approach proves robust across brain regions and conditions, it could make large-scale connectivity mapping accessible to individual labs rather than requiring consortium-level investment.
The discovery of previously unknown connections in a well-studied circuit also underscores how much remains to be learned about basic brain wiring. The pontocerebellar pathway is not obscure; it has been studied for decades. Finding new cell-type connections there suggests that other circuits likely harbor similar surprises.
The research was supported by the Wu Tsai Neurosciences Institute, the Elsa U. Pardee Foundation, and the Edward Mallinckrodt Jr. Foundation, according to the University of Illinois.