Artificial intelligence has revolutionized the way scientists map the brain, with a recent study producing one of the most detailed maps of the mouse brain to date. Comparing their AI program to ChatGPT, researchers have identified 1,300 regions and subregions on the map, some of which have never been charted before.
Led by researchers from the University of California at San Francisco and the Allen Institute, the study published in Nature Communications showcases the power of AI in understanding the complex structure of the brain. Senior author Reza Abbasi-Asl highlights the use of a ‘transformer’ network, similar to AI tools like ChatGPT, to analyze and interpret vast amounts of data.
Bosiljka Tasic, director of molecular genetics at the Allen Institute, emphasizes the importance of mapping the brain’s geography for advancing treatments for neurological ailments. With a detailed understanding of brain regions and their functions, researchers can develop more targeted drug therapies with fewer side effects.
The AI model used in the study, CellTransformer, analyzed spatial transcriptomics data from millions of brain cells to define neighborhoods within the brain. Starting with 25 regions, researchers gradually increased the resolution to identify 1,300 regions and subregions, uncovering previously unknown areas in the brain.
Tasic describes the process as going from a map of continents to one that includes states, cities, and neighborhoods within cities. By identifying neighbors of individual cells and grouping them into regions, CellTransformer has provided a comprehensive view of the mouse brain’s cellular structure.
The study opens up new possibilities for further research, including the potential for more detailed brain maps and the integration of different types of brain data. Tasic envisions a future where AI models can map the human brain with similar precision, although she acknowledges that data collection remains a significant challenge due to the brain’s complexity and size.
Overall, the AI-driven approach to brain mapping represents a significant step forward in neuroscience, offering new insights into the brain’s intricate architecture and paving the way for more targeted and effective treatments for neurological disorders.
It’s difficult to predict the exact timeline, but researchers suggest that it may take around ten more years to gather comprehensive data on the human brain similar to what has been achieved with the mouse brain.
Lead by UCSF scientist Alex Lee, the groundbreaking Nature Communications study sheds light on the intricacies of the brain with their research article titled “Data-Driven Fine-Grained Region Discovery in the Mouse Brain With Transformers.” Collaborating with experts like Alma Dubuc, Michael Kunst, Shenqin Lao, Nicholas Lusk, Lydia Ng, Hongkui Zeng, Bosiljka Tasic, and Reza Abbasi-Asl.
