ClusterMap for multi-scale clustering analysis of spatial gene expression

Abstract

Quantifying RNAs in their spatial context is crucial to understanding gene expression and regulation in complex tissues. In situ transcriptomic methods generate spatially resolved RNA profiles in intact tissues. However, there is a lack of a unified computational framework for integrative analysis of in situ transcriptomic data. Here, we introduce an unsupervised and annotation-free framework, termed ClusterMap, which incorporates the physical location and gene identity of RNAs, formulates the task as a point pattern analysis problem, and identifies biologically meaningful structures by density peak clustering (DPC). Specifically, ClusterMap precisely clusters RNAs into subcellular structures, cell bodies, and tissue regions in both two- and three-dimensional space, and performs consistently on diverse tissue types, including mouse brain, placenta, gut, and human cardiac organoids. We demonstrate ClusterMap to be broadly applicable to various in situ transcriptomic measurements to uncover gene expression patterns, cell niche, and tissue organization principles from images with high-dimensional transcriptomic profiles.

Spatially resolved transcriptomics provide a new method for cancer research

“…Consequently, novel algorithms are developed. For ISS methods, He et al [ 38 ] introduced an unsupervised and annotation-free framework, termed ClusterMap, which defined the task as a point pattern, and found significant biological structures by density peak clustering (DPC). Cable et al [ 39 ] developed robust cell type decomposition (RCTD), which could spatially map cell types and thus defined the spatial components of cell identity in Slide-seq and Visium datasets of the mouse brain.…”

Spatial Statistics for Understanding Tissue Organization

“…Describing local neighborhoods as vectors of counts of object types has been suggested in several publications under multiple names (Stoltzfus et al, 2020;He et al, 2021;Salas et al, 2021). Here we refer to it as the vector approach.…”

“…A neighborhood is a composition of object types inside a fixed area. The neighborhoods' locations can be uniformly allocated in a grid pattern throughout the space, constructed around each object from the dataset (Stoltzfus et al, 2020), based on Density peak clustering (He et al, 2021), or be defined by previously segmented tissue structures (Salas et al, 2021). Next, each neighborhood is presented as a vector containing counts of object types normalized, for example, by dividing each object count by the sum of all counts in the neighborhood (local normalization) or by dividing each object count by the sum of all the counts in the sample (global normalization).…”

“…The recently published Matisse(Salas et al, 2021) includes the Neighborhood enrichment test and the Vector approach. The toolbox histoCAT(Schapiro et al, 2017) includes the Neighborhood enrichment test, and Clustermap(He et al, 2021) includes the Vector approach. Table1…”

ReDisX: a Continuous Max Flow-based framework to redefine the diagnosis of diseases based on identified patterns of genomic signatures

“…Transforming the conventional diagnosis strategy for diseases is emerging for ensuring a healthier lifespan (Zhao et al, 2020) and for strengthening drug repurposing (Pushpakom et al, 2019). In the modern era, the practices for diagnosing diseases have been progressively oriented towards precision and personalized, relying on a molecular genetic basis, especially gene expression-based (Aure et al, 2017;He et al, 2021). Since the conventional diagnosis of diseases often remains insufficient in explaining heterogeneity within a disease and the homogeneity between multiple diseases (Humby et al, 2019;Khera & Kathiresan, 2017).…”