Beyond the nuclear border: single-cell analysis of in situ sequenced human brain tissue using cellular features

Area of Science:

• Neuroscience
• Genomics
• Biotechnology

Background:

• Spatial transcriptomics enhances understanding of cellular heterogeneity.
• Analyzing formalin-fixed, paraffin-embedded (FFPE) postmortem human brain tissue presents challenges for in situ sequencing (ISS).
• Optimizing transcript allocation is crucial for accurate analysis in complex tissues like the central nervous system (CNS).

Purpose of the Study:

• To assess the suitability of in situ sequencing (ISS) for FFPE postmortem human brain tissue.
• To compare geospatial methods for transcript allocation using nuclear and expanded nuclear boundaries.
• To develop and validate an enhanced method for targeted transcript allocation using integrated imaging data.

Main Methods:

• Comparison of geospatial segmentation and transcript allocation methods (nuclear vs. expanded nuclear boundaries).
• Integration of fluorescent imaging data (18S RNA and IBA1 protein) to guide transcript allocation.
• Development of a secondary allocation strategy for transcripts outside imaging masks.

Main Results:

• While overall cell-type proportions were similar, allocation methods impacted specific cell types like microglia, neurons, and neurovascular cells.
• Integrated imaging data improved the specificity of transcript allocation towards RNA-rich cells and microglia.
• The enhanced method increased the detection of microglia and the specificity of microglial transcript assignments.

Conclusions:

• The developed method offers a flexible and efficient strategy for targeted transcript allocation in FFPE human brain tissue.
• Integrating imaging data with ISS improves CNS cell segmentation and transcript specificity.
• This approach optimizes the analysis of spatial transcriptomics in challenging FFPE samples.