Scientists have used deep learning to analyze gene regulation in human, chicken, and mouse brains, revealing conserved and diverged cell type codes. The study provides new insights into brain evolution and offers tools for studying how gene regulation influences cell type development across species.
The complexity of the human brain is made up of many different types of cells, each with its own unique shape and function. These cell types are controlled by genetic switches that regulate which genes are turned on or off. To understand this complex puzzle, researchers have been working to decipher the regulatory code that makes these switches work.
Deep learning models, developed by a Belgian research team led by Prof. Stein Aerts, were used to analyze gene regulation data from thousands of individual cells. The models helped identify regulatory mechanisms across different cell types and explored whether the existing differences in brain anatomy between mammals and birds are reflected in shared or divergent regulatory codes.
The study found that some regulatory cell type codes are highly conserved between birds and mammals, while others have evolved differently. For example, the regulatory codes for certain bird neurons resemble those of deep-layer neurons in mammalian brains. The team also developed a comprehensive transcriptomic atlas to better understand the cell type composition of the chicken brain.
These findings offer new insights into brain evolution and provide valuable tools for studying how gene regulation influences cell type development across species. The study’s results have implications beyond understanding evolution, including identifying variants in genomes associated with mental or cognitive traits and disorders.
The research team is already applying their models to study the impact of genomic variants on disease, including Parkinson’s disease. They are also expanding their evolutionary modeling to many more animal brains, including fish, hedgehogs, and capibaras.
This groundbreaking study demonstrates the power of deep learning in understanding complex biological systems and has significant implications for the field of neuroscience and beyond.
Source: https://scitechdaily.com/ai-cracks-the-brains-genetic-code-unlocking-evolutionary-secrets