SM3DD with segmented PCA: a comprehensive method for interpreting 3D spatial transcriptomics

Area of Science:

• Spatial transcriptomics
• Computational biology
• Pathology

Background:

• Acute respiratory distress syndrome (ARDS) is a severe complication of SARS-CoV-2 infection.
• Understanding spatial gene expression in lung tissue is crucial for disease mechanism discovery.

Purpose of the Study:

• To develop a novel, cell segmentation-free method for analyzing spatial RNA datasets.
• To compare spatial gene expression patterns between normal lung tissue and SARS-CoV-2 infected lung tissue.

Main Methods:

• Developed Standardised Minimum 3D Distance (SM3DD) for spatial RNA analysis.
• Utilized CosMx™ Spatial Molecular Imager for RNA spatial coordinate determination.
• Applied hierarchical clustering and segmented principal components analysis to SM3DD data.

Main Results:

• SM3DD successfully identified differences in spatial gene expression between normal and SARS-CoV-2 lung tissue.
• Hierarchical clustering organized genes by functionality, aiding biological interpretation.
• Identified significant differences for FKBP11 and MZT2A, suggesting a role in interferon signaling.
• Detected pathways related to 'SARS-CoV-2 infection' without direct viral transcript detection.

Conclusions:

• SM3DD is an effective tool for analyzing spatial RNA data without cell segmentation.
• Spatial gene expression alterations in SARS-CoV-2 infection are identifiable using SM3DD.
• The method offers insights into disease mechanisms and potential therapeutic targets.