Spatial Transcriptomics (ST) Technologies and a Novel Solution: SPACEL
Spatial transcriptomics involves examining tissues by analyzing gene expression levels in individual cells, offering insights into cellular spatial organization and function.
To overcome the challenges presented by analyzing multiple tissue slices and the low resolution of ST spots, Prof. Qu Kun and his team pioneered Spatial Architecture Characterization by Deep Learning (SPACEL). This toolkit includes three modules—Spoint, Splane, and Scube—that work together to automatically create a 3D panorama of tissues.
Sprint, Splane, and Scube are each targeted at specific tasks, and employ innovative methods to enhance analysis and overcome challenges. By applying SPACEL to 11 ST datasets with various technologies, the researchers found that SPACEL outperformed previous techniques in several fundamental analytical tasks—cell type distribution prediction, spatial domain identification, and three-dimensional tissue reconstruction.
In conclusion, SPACEL introduces a significant step in spatial transcriptomics, enabling precise cell type predictions, effective spatial domain identification, and accurate 3D tissue alignment. It is a powerful tool to overcome the challenges associated with the joint analysis of multiple ST slices.
For more information, check out the research paper here.
If you love our work, join our ML Reddit, Facebook Community, Discord Channel, LinkedIn Group, Twitter, and Email Newsletter for the latest AI research news, cool AI projects, and more.
Rachit Ranjan is a consulting intern at MarktechPost currently pursuing his B.Tech from the Indian Institute of Technology (IIT) Patna, shaping his career in Artificial Intelligence and Data Science.