METI: deep profiling of tumor ecosystems by integrating cell morphology and spatial transcriptomics

Area of Science:

• Oncology
• Bioinformatics
• Genomics

Background:

• Spatial transcriptomics (ST) offers insights into tumor microenvironment (TME) cellular interactions.
• Existing ST analytical tools often neglect histological features and require matched single-cell RNA sequencing data, limiting their TME application.
• There is a need for integrated analytical frameworks that leverage morphological and transcriptomic data for comprehensive TME analysis.

Purpose of the Study:

• To introduce the Morphology-Enhanced Spatial Transcriptome Analysis Integrator (METI), an end-to-end framework for analyzing spatial transcriptomics data.
• To enhance the understanding of molecular and cellular landscapes within the TME by integrating ST, cell morphology, and gene signatures.
• To provide a robust tool for mapping cancer cells, TME components, stratifying cell types/states, and analyzing cell co-localization.

Main Methods:

• Developed METI, an integrated framework combining spatial transcriptomics, cell morphology, and curated gene signatures.
• Applied METI to ST data from diverse tumor tissues, including gastric, lung, and bladder cancers, and premalignant tissues.
• Performed quantitative comparisons of METI against established clustering and cell deconvolution methods.

Main Results:

• METI successfully maps cancer cells and TME components, stratifies cell types and states, and analyzes cell co-localization.
• Integration of morphological and transcriptomic data by METI provides enhanced insights into tissue-level cellular interactions.
• METI demonstrated robust and consistent performance across various cancer types and compared favorably to existing tools.

Conclusions:

• METI offers a powerful, integrated approach for analyzing spatial transcriptomics data in the context of tumor microenvironments.
• The framework advances the study of cellular interactions and molecular landscapes within complex tissues.
• METI's ability to incorporate histological features and gene signatures provides a more comprehensive understanding of cancer biology.