Abstract 6293: Developing and deploying the UCSC Xena gene set enrichment analysis appyter

Abstract UCSC Xena (https://xena.ucsc.edu) is a popular web-based visual integration and exploration tool for multiomic data and associated clinical and phenotypic annotations. Xena users want to be able to run Gene Set Enrichment Analysis (GSEA) to compare gene expression between their two dynamically created sample subgroups and determine which gene programs are upregulated or downregulated. However, the traditional GSEA implementation was not performant enough to support Xena’s interactive web-based data visualizations. To overcome this challenge, blitzGSEA, a more performant implementation of GSEA developed in 2022 by the Ma’ayan Lab at Mount Sinai, was integrated to be available within the Xena browser using the Appyter framework, a tool also developed by the Ma’ayan lab. Appyter allows users to use Jupyter Notebooks to run bioinformatic methods without having to interact with the code directly and also allows developers to dynamically code inputs, such as the subgroups created by a user. The framework handles users running analyses concurrently and will cache results if two users run the same analysis, such as part of a workshop. Before development, blitzGSEA was rigorously compared to traditional GSEA to confirm that it provided similar, biologically consistent results with efficient memory usage and runtime. The blitzGSEA Appyter was implemented alongside the addition of a Plotly Dash page to provide users with interactive plots and tables of the analysis results. The blitzGSEA Appyter was then deployed on the production Xena Browser using Docker and an AWS EC2 instance. The blitzGSEA Appyter is now publicly available to Xena users for any sample subgroups they create and averages 500 analyses per month around the world. Citation Format: Cally Lin, Jingchun Zhu, Mary Goldman. Developing and deploying the UCSC Xena gene set enrichment analysis appyter [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6293.