蛋白质组学,人工智能 z-bio 2025-02-11 271

　　Schematic diagram of high-resolution spatial proteome detection and analysis process at the level of whole tissue slices

　　Supported by the National Natural Science Foundation of China (Grant No.: 32025009, 32130020, 32400533, 32300538), Zhao Fangqing and Ji Peifeng from the Institute of Zoology of the Chinese Academy of Sciences have made progress in the field of artificial intelligence driven space proteomics technology. The research results are entitled "High resolution spatially resolved proteins of complex tissues based on microfluidics and transfer learning" and published online in Cell on January 23, 2025) Magazine, paper link: https://www.cell.com/cell/fulltext/S0092-8674 (24)01436-3#sec-5。

　　Space omics technology has become an important tool for analyzing tissue heterogeneity and complex cellular interaction mechanisms. Especially, spatial transcriptomics technology has shown great potential in embryonic development, neuroscience, and disease mechanism research. However, existing spatial proteomics techniques are limited in their application in complex tissue research due to factors such as mass spectrometry detection flux and high cost, making it difficult to meet the requirements of high-resolution and large-area tissue analysis.

　　The research team proposed PLATO, an artificial intelligence driven spatial proteomics measurement and analysis technology framework, which integrates deep learning algorithms with microfluidic technology to achieve high-resolution spatial proteome detection at the tissue slice level. Firstly, this study combined microfluidic technology to develop a high-throughput, low-cost parallel in-situ sampling platform that can achieve flexible and accurate sampling in the resolution range of 25 to 100 micrometers. Secondly, this study uses artificial intelligence algorithms to restore the spatial position information of proteins, breaking through the limitations of traditional mass spectrometry technology in obtaining spatial information and significantly improving the coverage and resolution of spatial proteomics. The research team used this technology to analyze the spatial distribution of proteins with high resolution in mouse brain tissue, intestinal villi, breast cancer and other complex tissues, further verifying the huge potential of this method in different application scenarios and research directions.

　　This study deeply integrates artificial intelligence algorithms, microfluidics, and mass spectrometry technologies, achieving important breakthroughs and technological iteration innovations in spatial omics technology. It provides strong support for revealing protein dynamic distribution and exploring the molecular mechanisms of complex biological processes, and is expected to play an important role in disease diagnosis, precision medicine, and agricultural biotechnology.