Visual analysis of spatial transcriptomics data with RedeViz

Abstract

Spatial transcriptomics (ST) technologies are powerful tools to illustrate the spatial hierarchy and heterogeneity of tissues with the lens of multiplexed gene readouts. However, ST technologies generate sequence data rather than images, preventing intuitive examination of the cellular contexture of tissues. Moreover, the inherent sparsity of ST data caused by molecular crowdedness and sequencing dropouts poses great challenges to accurate and clear visualization. In this study, we introduce RedeViz, a toolkit crafted for enhancing and visualizing subcellular-resolution ST data. RedeViz applies a pixel-level enhancement strategy, visualizes ST data in automatic or customized manners, and can display the cellular and genic spatial patterns with effects akin to HE staining. Strict evaluations confirm that RedeViz fits a wide range of ST platforms, including Xenium, Visium HD, MERFISH, CosMx, Stereoseq, as well as spatial proteomic platforms like CODEX. The impressive performance of RedeViz across various scales from cell-, tissue-, organ-, to organism-levels brings us a universal "What You See Is What You Get" framework for visual analysis of ST data.