Kyushu University researchers introduced QDyeFinder, an AI tool that enhances neuron mapping in the brain using advanced color coding and machine learning, showing potential for broader applications in biology.
Researchers have developed QDyeFinder, an AI pipeline that can entangle and reconstruct the brain's dense neuronal networks.
The brain is the most complex organ ever created. Its functions are supported by a network of tens of billions of dense neurons, in which trillions of connections exchange information and perform calculations. Trying to understand the complexity of the mind can be dizzying. Nevertheless, if we ever hope to understand how the brain works, we need to be able to map neurons and study their wiring.
Now, I'm getting published. Nature CommunicationsResearchers at Kyushu University have developed a new AI tool, which they call QDyeFinder, that can automatically identify and reconstruct individual neurons from images of mouse brains. The process involves tagging neurons with a super-multicolor labeling protocol, and then letting AI automatically identify the neuron's structure by matching combinations of similar colors.
Credit: Kyushu University/Takeshi Amai
Challenges in Neuron Mapping
“One of the biggest challenges in neuroscience is trying to map the brain and its connections. However, because neurons are so dense, it is very difficult and time-consuming to separate neurons from their axons and dendrites – they extensions that send and receive information from other neurons—to each other,” explains graduate school professor Takeshi Ami. of Medical Sciences, who led the study. “To put this in perspective, the axons and dendrites are only one micrometer thick, 100 times thinner than a standard human hair strand, and have less space between them.”
One strategy for identifying neurons is to tag the cell with a fluorescent protein of a specific color. Researchers can then trace the dye and reconstruct the neuron and its axons. By expanding the color range, more neurons can be detected simultaneously. In 2018, Imai and his team developed Tetbow, a system that can make neurons glow with three primary colors of light.
Mouse cortical layer 2/3 pyramidal neurons were labeled with 7-color Tetbo. A combination of 7 fluorescent proteins (mTagBFP2, mTurquoise2, mAmetrine1.1, mNeonGreen, Ypet, mRuby3, tdKatushka2) was used to visualize the dense wiring of neurons. The 7-channel images were then analyzed by the QDyeFinder program to reveal the wiring patterns of individual neurons. Credit: Kyushu University/Takeshi Amai
“An example I like to use is a map of the Tokyo subway lines. The system spans 13 lines, 286 stations and over 300 kilometers. Each line on the subway map is color-coded, this So you can easily identify which stations are connected,” explains Markus N. Levy, first author of the paper and then assistant professor. “Tetbo detected neurons and their connections Made searching very easy.”
However, two important problems remained. Neurons still had to be traced carefully by hand, and using only three colors was not enough to determine a large population of neurons.
Technical breakthroughs with QDyeFinder
The team worked to increase the number of colors from three to seven, but then the biggest problem was the limitations of human color perception. Look closely at any TV screen and you'll see that pixels are made up of three colors: blue, green, and red. Any color we can see is a combination of these three colors, just as our eyes have blue, green and red sensors.
“Machines, on the other hand, have no such restrictions. So, we worked to develop a tool that can automatically distinguish these vast color combinations,” Leiwe continues. “We also designed it so that the tool would automatically group neurons and axons of the same color together and reconstruct their structure. We called the system QDyeFinder.
QDyeFinder works by first automatically identifying fragments of axons and dendrites in a given sample. It then identifies the color information of each fragment. Then, using a machine learning algorithm the team developed called dCrawler, the color information was grouped together, identifying the axons and dendrites of the same neuron.
“They had the same results when we compared the QDyeFinder results to data from manually traced neurons. correctionLevi explained. “Even compared to existing tracing software that uses full Machine learningQDyeFinder was able to identify axes with much greater accuracy.”
The team hopes their new device can advance the ongoing quest to map the brain's connections. They would also like to see if their new method can be applied to labeling and tracking other complex cells such as cancer cells and immune cells.
“The day may come when we can read the connections in the brain and understand what they mean or represent the person. I doubt it will happen in my lifetime, but our work is ours. represents a solid step forward in understanding perhaps the most complex and mysterious dimension of existence,” Amy concluded.
Citation: “Automated Neuronal Reconstruction and Color Threshold-Based Clustering with Super-Multicolor Tetbo Labeling” Marcus N. Levy, Satoshi Fujimoto, Toshikazu Baba, Daiichi Moriasu, Baswanath Saha, Ritchie Sakaguchi, Shigenori 2 June 2024, Ami Nature Communications.
DOI: 10.1038/s41467-024-49455-y
Funding: Japan Agency for Medical Research and Development, Japan Science and Technology Agency, Japan Society for Promotion of Science, Uehara Memorial Foundation, Sumitomo Foundation, Ichiro Kinhara Foundation, Daiichi Sankyo Foundation of Life Science, Brain Science Foundation