Robust decomposition of cell type mixtures in spatial transcriptomics

Dylan M. Cable,Evan Murray,Luli S. Zou,Aleksandrina Goeva,Evan Z. Macosko,Fei Chen,Rafael A. Irizarry
DOI: https://doi.org/10.1038/s41587-021-00830-w
IF: 46.9
2021-02-18
Nature Biotechnology
Abstract:A limitation of spatial transcriptomics technologies is that individual measurements may contain contributions from multiple cells, hindering the discovery of cell-type-specific spatial patterns of localization and expression. Here, we develop robust cell type decomposition (RCTD), a computational method that leverages cell type profiles learned from single-cell RNA-seq to decompose cell type mixtures while correcting for differences across sequencing technologies. We demonstrate the ability of RCTD to detect mixtures and identify cell types on simulated datasets. Furthermore, RCTD accurately reproduces known cell type and subtype localization patterns in Slide-seq and Visium datasets of the mouse brain. Finally, we show how RCTD's recovery of cell type localization enables the discovery of genes within a cell type whose expression depends on spatial environment. Spatial mapping of cell types with RCTD enables the spatial components of cellular identity to be defined, uncovering new principles of cellular organization in biological tissue. RCTD is publicly available as an open-source R package at <a href="https://github.com/dmcable/RCTD">https://github.com/dmcable/RCTD</a>.
biotechnology & applied microbiology
What problem does this paper attempt to address?
The paper aims to address the problem of cell type mixture identification and decomposition in spatial transcriptomics technology. Specifically, the paper proposes a computational method called RCTD (Robust Cell Type Decomposition), which utilizes single-cell RNA sequencing (scRNA-seq) data to decompose cell type mixtures in spatial transcriptomics data and corrects for differences between different sequencing technologies. ### Main Issues: 1. **Challenges of Cell Type Mixtures**: In spatial transcriptomics, a single measurement point may contain information from multiple cells, making it difficult to discover the localization and expression patterns of specific cell types. 2. **Technical Platform Effects**: There are systematic technical differences between different sequencing technologies, which affect the detection rate of gene expression and thus impact the accuracy of cell type identification. ### Contributions of the Paper: - Proposed the RCTD method, which can decompose individual cell types from spatial transcriptomics data and perform cross-platform learning between different sequencing technologies. - RCTD estimates the proportion of cell types at each pixel point through statistical modeling and considers the impact of platform effects. - Validated the effectiveness of RCTD on both simulated and real datasets, including Slide-seq and Visium datasets of the mouse brain. - Demonstrated how RCTD can reveal the spatial components of cell identity through spatial mapping of cell types, thereby discovering new biological tissue principles. ### Conclusion: RCTD can accurately identify and decompose cell type mixtures in spatial transcriptomics data and correct for differences between different sequencing technologies, thereby improving the understanding of cell type localization and expression patterns.