Machine learning method improves understanding of cell identity — Zoo House News
When genes are activated and expressed, they show patterns in cells that are similar in type and function in all tissues and organs. The discovery of these patterns improves our understanding of cells – which has implications for unveiling disease mechanisms.
The advent of spatial transcriptomics technologies has enabled researchers to observe gene expression in its spatial context across whole tissue samples. However, new computational methods are needed to understand this data and to identify and understand these gene expression patterns.
A research team led by Jian Ma, Ray and Stephanie Lane Professor of Computational Biology in Carnegie Mellon University’s School of Computer Science, has developed a machine learning tool to fill this gap. Her article on the method, called SPICEMIX, made the cover story in the latest issue of Nature Genetics.
SPICEMIX is helping researchers unravel the role that different spatial patterns play in overall gene expression by cells in complex tissues such as the brain. To do this, each pattern is represented with spatial metagenes—groups of genes that may be associated with a specific biological process, and that may show smooth or sporadic patterns in the tissue.
The team Ma was on; Benjamin Chidester, project scientist in the Computational Biology department; and Ph.D. Students Tianming Zhou and Shahul Alam used SPICEMIX to analyze spatial transcriptomics data from brain regions in mice and humans. They used the unique capabilities of SPICEMIX to uncover the landscape of the brain’s cell types and spatial patterns.
“When choosing the name, we were inspired by cooking,” said Chidester. “You can make all sorts of flavors with the same spices. Cells can function in a similar way. They may use a common set of biological processes, but the specific combination they use gives them their unique identity.”
When applied to brain tissue, SPICEMIX identified spatial patterns of cell types in the brain more accurately than other methods. It also uncovered new expression patterns of brain cell types through the learned spatial metagenes.
“These results can help us paint a more complete picture of the complexity of brain cell types,” Zhou said.
The number of studies using spatial transcriptomics technologies is growing rapidly, and SPICEMIX can help researchers make the most of this high-volume, high-dimensional data.
“Our method has the potential to advance spatial transcriptome research and contribute to a deeper understanding of both basic biology and disease progression in complex tissues,” said Ma.