PreDigs: a Database of Context-specific Cell-type Markers and Precise cell subtypes for Digestive Cell Annotation

Abstract

Research on cell type markers aids investigators in exploring the diverse cellular compositions within gastrointestinal tumors, enhancing our understanding of tumor heterogeneity and its implications for disease progression and treatment response. However, issues such as the integration of large-scale datasets and the lack of standardized cell type identification hinder comprehensive characterization. Here, we developed a user-friendly web interface called PreDigs (Predicted Signatures in Digestive System), which offers 124 tailored scRNA-seq datasets available for download, encompassing over 3.4 million cells. After unsupervised clustering, we unified the identification and naming of subtype labels, ultimately constructing a cell ontology tree that includes 142 cell types, with up to eight hierarchical levels. Meanwhile, we calculated three different context-specific cell-type markers--Cell Markers, Subtype Markers, and TPN Markers--based on various application requirements within or across tissues. Through the integrated analysis of PreDigs gastrointestinal data, we identified distinct cell subpopulations exclusive to tumors, one of which corresponds to tumor-specific endothelial cells (TEC). Furthermore, PreDigs offers online cell annotation tools that empower users to perform single-cell classification with greater flexibility, accessible at https://www.biosino.org/predigs/.