Janelia Research Campus is tripling its fish lab space and projecting a tenfold team expansion to use the tiny transparent Danionella to watch a vertebrate brain in action as the animal behaves.
A fish no longer than a fingernail sits under a microscope, its skull and skin see-through, the neurons in its brain lit up by fluorescent proteins. For the first time, researchers can watch a vertebrate brain in action while the animal still swims, eats, and reacts. That image is the centerpiece of a major institutional wager at one of the world's leading neuroscience centers.
HHMI's Janelia Research Campus, a Howard Hughes Medical Institute facility in Ashburn, Virginia, about 35 miles from Washington, D.C., built much of its reputation on the fruit fly Drosophila. The campus is now pivoting toward Danionella, a tiny transparent fish, betting that watching a vertebrate brain work in real time will finally let researchers connect neural activity to behavior in a living animal.
The bet is concrete. Janelia is tripling the space dedicated to Danionella, with new lab space expected to reach roughly 6,000 square feet. The team of scientists working with the fish is projected to grow from about 10 to 100 or more, a tenfold expansion. That scale signals a strategic shift, not a side project, and NPR's reporting on June 16, 2026, by science correspondent Jon Hamilton, frames the move as "a huge gamble" by "one of the world's top centers for brain science".
The scientific problem the wager tries to solve is older than the building. The brain is hidden inside a skull, and even the most powerful microscopes have struggled to capture the activity of every neuron in a living animal while the animal behaves. Fruit flies, the workhorse of Janelia's connectome project, are transparent as larvae, but their adult brains are dense, and the evolutionary distance to humans is large. Mice are mammals, but their brains are opaque. Danionella is small enough, transparent enough, and genetically close enough to humans to sit in a useful middle: scientists can see into its brain, and the circuitry that runs its behavior is in many cases the same circuitry that evolution carried forward into us. The lab's stated goal is to understand how the brain controls behavior in animals and ultimately in humans.
Fluorescent proteins, the same family of tools that earned their discoverers a Nobel Prize, are what make the see-through trick work. Modified versions of those proteins light up specific cell types, so when researchers peer into a Danionella, they can watch entire brain circuits fire as the fish responds to its environment, all in real time. The accompanying research imagery, credited to Chie Satou of HHMI, captures exactly that: a fish, a microscope, and a brain that can be read while it runs.
AI is the method that makes the volume of that data usable. A whole brain of a living fish, imaged at single-cell resolution while the animal moves, produces a torrent of activity. Machine learning tools, the same kind that drove the recent wave of protein-structure prediction and image segmentation, are now being tuned to track individual neurons across frames, classify behaviors, and map the relationship between firing patterns and actions. Without AI, the data would be too large to interpret. With it, the lab gets the read it has been waiting for: brain and behavior in the same animal, at the same time.
Janelia is not declaring fruit flies or mice obsolete. Established model organisms still have strengths flies and mice do not, and the move comes with a real cost: a generation of expertise, tools, and published results built around Drosophila. That is the gamble. If whole-brain imaging in a living vertebrate delivers the answers the field has chased, the wager pays off. If the fish turns out to be a quirky model whose lessons do not transfer, the redirected resources and rebuilt expertise will be visible for years.
What to watch next: the first peer-reviewed papers out of the new Danionella facility, the scientists Janelia hires in the next round of expansion, and whether other major neuroscience centers follow with their own transparent-vertebrate programs. The fish is the tool. The story is whether 6,000 square feet, a tenfold team, and a see-through vertebrate are enough to crack the brain-behavior code that fruit flies could not.