Abstract 7433: CGC-hosted comprehensive pipeline for spatial transcriptomics analysis

Abstract

Abstract Spatial transcriptomics has been experiencing significant growth in the past years. Inspired by two widely adopted methodologies, particularly in situ hybridization and next-generation sequencing, with an emphasis on the single-cell RNA sequencing (scRNA-seq), this hybrid method facilitates whole transcriptome profiling while preserving spatial context at high resolutions, thus allowing the exploration of novel insights in cancer research. We introduce the highly configurable solution for sequencing-based technologies that allow comprehensive spatial analysis. The pipeline is available on NCI-funded Cancer Genomics Cloud (CGC) platform, powered by Seven Bridges. The CGC platform provides a collaborative cloud base computation infrastructure that collocates computation, over 1000 bioinformatics workflows, and 4+ PB data to researchers, making the analysis of the Cancer Research data Commons (CRDC) datasets accessible from any environment. This pipeline is developed using reliable, widely adopted packages and can process datasets generated by several leading technologies. It comprises a set of common steps including quality control, data preprocessing, dimensionality reduction, cluster identification, detection of spatially variable features, and integration with scRNA-Seq reference. The pipeline is highly configurable, allowing for various settings to be modified and optimized for result refinement. Additionally, it allows the selection of only specific components of the pipeline to be executed. The key steps of the pipeline are visually represented, allowing researchers to get a detailed insight into results. Here, we demonstrate a typical flow of spatial transcriptomics analysis on publicly available datasets using the developed pipeline. We illustrate the impact of pipeline settings on the analysis outcomes. Furthermore, we identify spatially variable genes whose expressions show a distinct localization within the tissue. Finally, we perform the data integration to predict the cell type composition within the determined spatial domains. Spatial transcriptomics analysis is a powerful method that has highly improved critical aspects of cancer research, such as the characterization of tumor microenvironment, the discovery of novel biomarkers and the clarification of drug resistance mechanisms. The ongoing evolution of this method is expected to play vital role in our deep understanding of complex spatial relationships within tissues. This CGC-hosted workflow is developed to contribute to the promising advancements. Citation Format: Miona Rankovic, Nevena Vukojicic, Nevena Ilic Raicevic, Vida Matovic, Divya Sain, Jack DiGiovanna, Brandi Davis-Dusenbery. CGC-hosted comprehensive pipeline for spatial transcriptomics analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7433.