Optimizing Xenium In Situ data utility by quality assessment and best-practice analysis workflows

Area of Science:

• Spatial transcriptomics
• Genomics
• Bioinformatics

Background:

• The Xenium In Situ platform offers in situ gene mapping at subcellular resolution.
• Choosing among diverse spatial transcriptomics technologies requires clear guidance.
• Independent performance evaluation is crucial for platform selection and data analysis.

Purpose of the Study:

• To independently evaluate the Xenium In Situ platform's performance across multiple datasets.
• To compare Xenium with other spatial transcriptomics technologies regarding scalability, resolution, and data quality.
• To benchmark open-source computational tools for Xenium data analysis and provide best practices.

Main Methods:

• Analysis of 25 Xenium In Situ datasets from various tissues and species.
• Comparative assessment against eight other spatial transcriptomics platforms.
• Benchmarking of open-source bioinformatics tools for preprocessing, cell segmentation, and feature selection.

Main Results:

• Detailed comparison of Xenium's scalability, resolution, data quality, and limitations.
• Performance metrics for various computational tools applied to Xenium data.
• Identification of optimal analysis workflows for Xenium datasets.

Conclusions:

• Xenium In Situ platform demonstrates specific capabilities and limitations.
• Recommendations for selecting spatial transcriptomics platforms and analysis strategies are provided.
• Best practices for analyzing Xenium data are established for the research community.