ASURAT

ASURAT performs integrated unsupervised clustering and functional annotation of single-cell RNA sequencing (scRNA-seq) and spatial transcriptome data to identify cell subpopulations and map genes to disease-, cell type-, biological process-, and signaling pathway-level functional terms.

Key Features:

• Unsupervised Clustering: ASURAT employs a correlation graph decomposition approach to cluster scRNA-seq and spatial transcriptome data and identify distinct cell subpopulations.
• Simultaneous Functional Annotation: The method annotates clusters with disease, cell type, biological process, and signaling pathway functional terms derived from biological databases.
• Gene-to-Term Mapping: It maps genes or gene modules onto database-derived functional terms via the correlation graph decomposition framework.
• Differential Expression Support: ASURAT aids identification of differentially expressed genes within clusters to support biological interpretation.

Scientific Applications:

• Identification of Subpopulations: Applied to scRNA-seq and spatial transcriptome datasets to detect previously overlooked subpopulations in human small cell lung cancer and pancreatic ductal adenocarcinoma.
• Gene Expression Analysis: Facilitates identification of differentially expressed genes to inform disease mechanisms and potential therapeutic targets.

Methodology:

ASURAT uses correlation graph decomposition to perform unsupervised clustering and maps genes onto functional terms derived from biological databases to enable simultaneous cluster annotation.