STARCH: copy number and clone inference from spatial transcriptomics data

Abstract

Tumors are highly heterogeneous, consisting of cell populations with both transcriptional and genetic diversity. These diverse cell populations are spatially organized within a tumor, creating a distinct tumor microenvironment. A new technology called spatial transcriptomics can measure spatial patterns of gene expression within a tissue by sequencing RNA transcripts from a grid of spots, each containing a small number of cells. In tumor cells, these gene expression patterns represent the combined contribution of regulatory mechanisms, which alter the rate at which a gene is transcribed, and genetic diversity, particularly copy number aberrations (CNAs) which alter the number of copies of a gene in the genome. CNAs are common in tumors and often promote cancer growth through upregulation of oncogenes or downregulation of tumor-suppressor genes. We introduce a new method STARCH (Spatial Transcriptomics Algorithm Reconstructing Copy-number Heterogeneity) to infer CNAs from spatial transcriptomics data. STARCH overcomes challenges in inferring CNAs from RNA-sequencing data by leveraging the observation that cells located nearby in a tumor are likely to share similar CNAs. We find that STARCH outperforms existing methods for inferring CNAs from RNA-sequencing data without incorporating spatial information.

Exploring tissue architecture using spatial transcriptomics

“…RNA Velocity 142 makes use of the unspliced transcripts to infer how spots are related to each other in time, and was applied in the cortex to map the dynamics of neuro-development 53 . RNA-Seq-based Copy-number variation inference identifies chromosomal aneuploidies, which can be used to distinguish malignant from non-malignant spots, and also identify distinct subclones 143,144 . When two sets of spots are spatially adjacent, potential modes of interaction 145 between the cells can be proposed by examining their paired receptors and ligands 111 using known databases such as CellPhoneDB 47,93,146 or NicheNet 147 .…”

Automated assignment of cell identity from single-cell multiplexed imaging and proteomic data

“…Although Astir is primarily designed for highly multiplexed imaging technologies, we do not currently make use of the spatial locations of cells for the purposes of cell type assignment. Future work may incorporate custom spatially-informed priors that have recently been incorporated into spatial clonal inference [29]. For example, in the context of breast cancers a cell adjacent to a large empty area interior to a duct (the lumen) is an epithelial cell.…”

Alignment and Integration of Spatial Transcriptomics Data

“…The aligned and integrated ST layers produced by PASTE can be used to increase the statistical power in multiple downstream analyses including: identification of spatial expression patterns [42, 30], spatial cell type annotation [5], tumor/normal spot classification [52], spatial cell-cell communication patterns [2, 9], identification of genomic copy number aberrations [17], and more. In addition, PASTE could be applied to ST experiments from different patients in order to find conserved patterns of gene expression across different patients.…”

“…We formalize the problem of finding a center ST layer by combining the ideas of fused Gromov-Wasserstein barycenter [46] and Non-Negative Matrix Factorization [28]. NMF has been shown to be useful in single-cell RNAseq analysis both as a method to impute missing values (“dropouts”) and as a dimensionality reduction technique [41, 53, 17]. In the C enter L ayer I ntegration P roblem similar to the fused Gromov-Wasserstein barycenter problem – we seek to find a center ST layer that minimizes the weighted sum of distances to a given set of input ST layers, where the distance between layers is calculate by the minimum value of the P airwise L ayer A lignment P roblem objective across all mappings.…”