Researchers at EPFL have figured out how to make nanopores behave like neurons — and it comes down to electrical charge.
A team led by Matteo Dal Peraro and Aleksandra Radenovic discovered that these microscopic tunnels can be engineered to "learn" from electrical signals, mimicking the way brain synapses strengthen or weaken over time. The breakthrough opens a path toward building computers that process information more like biological systems do.
Here's what makes this possible: nanopores are tiny holes in membranes, and ions naturally flow through them. But the team found that by controlling the electrical charges lining the pore's interior, they could make the pores behave in specific ways — either favoring ion flow in one direction (rectification) or temporarily closing in response to strong electrical signals (gating).
We're a new kind of news feed.
Regular news is designed to drain you. We're a non-profit built to restore you. Every story we publish is scored for impact, progress, and hope.
Start Your News DetoxTo understand this, the researchers engineered 26 different versions of a bacterial pore called aerolysin, each with a different arrangement of charged amino acids. They then watched how ions moved through each variant under different conditions. The pattern became clear: the pore's internal charge distribution acts like a gatekeeper, determining how easily ions can pass and when the structure becomes unstable enough to shut down temporarily.
The flexibility of the pore itself turned out to be crucial. When the team made pores more rigid, gating stopped entirely — the pore lost its ability to respond dynamically. This suggests that structural flexibility, like the adaptability of real neural tissue, is essential for learning-like behavior.
What makes this practically important is that engineers can now deliberately design nanopores for different jobs. Want a sensor that won't accidentally close? Adjust the charges to eliminate gating. Need a component for bio-inspired computing? Build in gating and watch it adapt to incoming signals, just like a synapse does.
The team demonstrated this by creating a nanopore that learned from repeated voltage pulses, strengthening its response over time — the same principle behind how brains form memories. This suggests that future processors built around ion channels could eventually perform computation in ways that are fundamentally different from silicon chips, potentially using far less energy in the process.
The research points toward a new class of adaptive, programmable materials that blur the line between hardware and biology.
