Fig. 1

Ten analytic modules and their hierarchical relationship in SRT-Server. SRT-Server features ten interconnected analytic modules (boxes), each containing one or more analytic methods (small font letters). Users can connect these modules in a hierarchical manner to form a customized analytic pipeline. The pipeline always begins with user-uploaded data in a generic or platform-specific format. Next, the QC module performs quality control and saves the processed data in h5 format. From the QC module, the pipeline branches into five analytic modules: the DECON module, which uses CARD to estimate cell type compositions on each measured spot through deconvolution; the DECON_PY module, which uses cell2location and tangram to perform deconvolution; the SVG module, which identifies spatial variable genes (SVGs) with SPARK or SPARK-X; the CT module, which clusters cells for single-cell resolution SRT using Seurat or BASS, or annotates them using scSorter or Garnett; and the SDD module, which clusters spatial domains using SpatialPCA or BASS. From the clustering modules (CT or SDD), the pipeline moves to three additional analytic modules: the TRAJ module for pseudo-time inference of cell types or spatial domains; the CCC module for detecting cell–cell or domain-domain interactions using a ligand-receptor database provided on the SRT-Server; and the DEG module for identifying differentially expressed genes (DEGs) based on the detected cell types or spatial domains. For genes output from the SVG or DEG modules, the ORA module estimates enrichment of biological pathways using five pathway databases provided on the SRT-Server. Note that regular rectangular boxes represent upstream analysis (procedures before CT), while rounded rectangular boxes represent downstream analysis based on clustering results