Protocol for inferring cell type relationships and linking genomic annotations to cell type hierarchies using CellWalker2

Single-cell multiomic data reveal cell type-specific regulatory mechanisms, but comparing hierarchies and regulatory programs across related cell types is challenging. Here, we present a protocol for integrating single-cell data to infer cell type relationships and label genomic annotations within cell type hierarchies using CellWalker2, a graph diffusion-based framework. We describe steps for cell labeling; comparing cell type hierarchies; mapping regulatory elements or transcription factors onto cell type hierarchies; and assessing statistical significance, using human blood and brain datasets. For complete details on the use and execution of this protocol, please refer to Hu et al.¹.