Democratizing the spatial view: STAMP technology from an analytical perspective

Area of Science:

• Molecular Biology
• Cell Biology
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) revolutionized cellular heterogeneity analysis but loses spatial, morphological, and protein localization data.
• Spatial transcriptomics offers partial context restoration but faces limitations in cost, throughput, and complexity.

Purpose of the Study:

• To introduce Single-cell Transcriptomics Analysis and Multimodal Profiling (STAMP) as a method to preserve cell visual characteristics alongside molecular data.
• To detail the downstream analytical workflow for STAMP datasets using Python.
• To demonstrate STAMP's utility in linking gene expression programs to cell morphology.

Main Methods:

• Immobilizing cells or nuclei on a slide for high-content imaging before molecular readout.
• Applying standard single-cell analysis methods to image-derived cell-by-feature matrices.
• Utilizing Python for quality control, normalization, dimensionality reduction, clustering, and interpretation.
• Implementing morpho-transcriptomic coupling to link gene programs with morphology.

Main Results:

• STAMP successfully preserves cell morphology and spatial context with high-plex transcript and/or protein measurements.
• Standard single-cell analysis pipelines are adaptable to STAMP data, yielding interpretable results.
• Morpho-transcriptomic coupling provides a strategy to connect gene expression patterns with visual cell phenotypes.

Conclusions:

• STAMP provides a practical bridge between single-cell transcriptomics and quantitative cell phenotype analysis.
• This method enables interpretable cell state inference that can be validated using imaging data.
• STAMP overcomes limitations of traditional scRNA-seq and spatial transcriptomics platforms.