Abstract 6309: CliPP-on-Web: a dynamic platform for clonal structure identification through penalizing pairwise differences in cancers

Background: Intra-tumor heterogeneity plays a crucial role in tumor evolution and response to therapy, highlighting the need to understand mutation clonality and subclonal architecture to drive advancements in precision cancer treatments. However, existing tools for subclonal reconstruction are typically too slow for large cohort analyses, and lack an accessible web-based interface for widespread use. Methods: We developed CliPP-on-Web, an interactive platform built with R Shiny, to streamline subclonal analysis for researchers and clinicians. Its back-end algorithm is based on model-based clustering with regularized maximum likelihood estimation, derived from the core methodology of CliPP. This algorithm models variant allele counts using a binomial distribution, incorporating tumor purity, copy number alterations (CNAs), and cancer cell fractions (CCFs) of somatic mutations. By penalizing pairwise differences in CCFs, it effectively clusters mutations without requiring predefined subclone numbers. Optimization is achieved through the Alternating Direction Method of Multipliers (ADMM), ensuring high computational efficiency. Results: CliPP-on-Web simplifies subclonal analysis by enabling users to process up to 10 samples simultaneously with user-uploaded. Interactive visualization tools, such as Variant Allele Frequency (VAF) and Cellular Prevalence (CP) plots, offer features like zooming, hover details, and adjustable smoothing for enhanced clarity. Pre-loaded integrated datasets from TCGA and PCAWG enable comparative studies, while the Driver Mutation module highlights key cancer mutations with searchable and downloadable results. Capable of processing 100 samples in just30 minutes — far outpacing the ∼3 hours per sample required by tools like PhyloWGS — CliPP-on-Web provides efficient and comprehensive subclonal analysis, advancing cancer genomics research. Conclusions: CliPP-on-Web offers a powerful, accessible platform for analyzing mutation clonality and subclonal architecture across diverse tumor types, facilitating both large-scale studies and individual sample exploration. Citation Format: Aaron Wu, Quang Tran, Matthew D. Montierth, Yujie Jiang, Jessica C. Lai, Shuangxi Ji, Wenyi Wang. CliPP-on-Web: a dynamic platform for clonal structure identification through penalizing pairwise differences in cancers [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 6309.