Deep learning and alignment of spatially-resolved single cell transcriptomes with Tangram

By Tommasso Biancalani, Gabriele Scalia, Lorenzo Buffoni, Raghav Avasthi, Aman Sanger, Neriman Tokcan, Charles R. Vanderburg, Å. Segerstolpe, Meng Zhang, I. Avraham-Davidi and 11 more. Nature Methods, 2021

721 citations recorded by the source corpus.

Charting an organs’ biological atlas requires us to spatially resolve the entire single-cell transcriptome, and to relate such cellular features to the anatomical scale. Single-cell and single-nucleus RNA-seq (sc/snRNA-seq) can profile cells comprehensively, but lose spatial information. Spatial transcriptomics allows for spatial measurements, but at lower resolution and with limited sensitivity. Targeted in situ technologies solve both issues, but are limited in gene throughput. To overcome these limitations we present Tangram, a method that aligns sc/snRNA-seq data to various forms of spatial data collected from the same region, including MERFISH, STARmap, smFISH, Spatial Transcriptomics (Visium) and histological images. Tangram can map any type of sc/snRNA-seq data, including multimodal data such as those from SHARE-seq, which we used to reveal spatial patterns of chromatin accessibility. We demonstrate Tangram on healthy mouse brain tissue, by reconstructing a genome-wide anatomically integrated spatial map at single-cell resolution of the visual and somatomotor areas. Tangram is a versatile tool for aligning single-cell and single-nucleus RNA-seq data to spatially resolved transcriptomics data using deep learning.

Subfields: Immunology, Molecular Biology, Neurology

View Paper: https://www.semanticscholar.org/paper/5152c6254aed0bb3c5fca2c9cfaf1fb3d52971bb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Immunology

In Molecular Biology

Citations in the corpus

Spatial components of molecular tissue biology (2022, positive): “For example, a recent method accounts for nuclei densities derived from tissue and adopts a probabilistic framework to assign cell-type proportions 116,143 .”
Benchmarking spatial and single-cell transcriptomics integration methods for transcript distribution prediction and cell type deconvolution (2022, methodology): “…the domain adaptation algorithm PRECISE 14 and k -nearest-neighbor regression to predict the spatial distribution of undetected transcripts; Tangram 15 uses non-convex optimization and a deep learning framework to learn a spatial alignment for scRNA-seq data; Seurat 16 applies canonical…”
Spatially informed clustering, integration, and deconvolution of spatial transcriptomics with GraphST (2023, positive): “Unlike Tangram 34 , which learns the mapping matrix directly using the raw gene expressions with high noise levels, GraphST retains and re ﬁ nes the informative and noise-reduced features from the raw gene expression pro ﬁ les of both the scRNA-seq and ST data via deep learning modules before…”
Spatiotemporal transcriptomic atlas of mouse organogenesis using DNA nanoball patterned arrays (2021, positive): “To validate our results, we aligned with scRNA-seq data using Tangram (Biancalani et al., 2021), which confirmed the reported probabilistic spatial profile for specific cell types such as choroid plexus, hindbrain glioblasts, forebrain GABAergic neuroblasts, midbrain glutamatergic neuroblasts, corti”
Best practices for single-cell analysis across modalities (2023, methodology): “GCNG 237 , Misty 238 and NCEM 236 formulate this task in terms of spatial graphs of cells and graph neural networks, SpaOTsc 239 uses optimal transport, and SVCA 240 quantifies the effect of cell–cell communication events on gene expression profiles with spatial variance component analysis.”
Methods and applications for single-cell and spatial multi-omics (2023, background): “single-cell data to identify the cellular constituents of a particular tissue region 149–151 .”