Enhancing Single Cell RNA-seq Analysis and Annotation for Understanding the Impact of CD137 Agonist on Cancer Immunoprevention through Advanced Informatics Algorithms

SA-4-1BBL, a novel CD137 receptor agonist, has been generated by our group and selected as a promising candidate for preventing cancer progression in research. The SA-4-1BBL demonstrates notable immunoprevention efficacy in diverse tumor models when utilized as a treatment in research subjects. Nevertheless, this efficacy stands in contrast to 3H3, another CD137-targeting antibody, which appears failing in generating cancer prevention outcomes in research subjects. The differing results obtained with SA-4-1BBL and 3H3 have raised questions about the reasons behind this disparity even though SA-4-1BBL and 3H3 target the same CD137 receptor. To uncover the underlying differences between SA-4-1BBL, 3H3, and the control (Naive, no treatment), our research group designed this research based on the mouse model to have three groups of mice with SA-4-1BBL compounds, 3H3 compounds, and no treatment. Each group of experimental subjects consists of five replicates, and single-cell RNA-sequencing (scRNA-seq) is performed on the lymph nodes of these experimental subjects. The scRNA-seq data analysis begins with data quality checking through FastQC and MultiQC tools. Subsequently, the Cell Ranger pipeline, developed by 10X Genomics, is utilized for read alignment, feature-barcode matrix calculations, and sample replicate aggregation with the GRCm39 mouse reference genome and a Gene Transfer Format (GTF) file from the Ensembl website. The outputs of the Cell Ranger pipeline are analyzed with Seurat for cluster identification. Besides that, our research group introduces a custom cell annotation approach, which not only identifies cell types but also dives into cell subtypes. This method relies on an annotation reference list composed of curated markers from published literature with a combination of prior knowledge of gradient-based expression levels for both general and specific markers within each cell type and subtype. Furthermore, the cell trajectory and pseudotime analysis are also accomplished with Slingshot, and the cell-cell communication analysis is also performed with the CellChat package. At the current stage of this research, our research group has identified 29 clusters of cells and utilized our cell annotation method to annotate cell types and subtypes. Besides that, cell trajectories, pseudotimes, and cell-cell communications of cell types and subtypes have also been generated for comparisons between different conditions. In future work, more detailed comparisons between conditions will be performed to uncover the underlying differences between the SA-4-1BBL and 3H3 compounds with the incorporation of bulk RNA-seq and flow cytometry data.