Abstract 2504: SNAP: Streamlined Nextflow Analysis Pipeline for profiling circulating histone modifications identifies tumor epigenomic signatures in cancer plasma

Abstract Background: Cell-free chromatin immunoprecipitation sequencing (cfChIP-seq) and methylated DNA immunoprecipitation sequencing (cfMeDIP-seq) are increasingly being used to profile cancer epigenomes from blood. The data these assays generate require specialized analytical approaches that are not met by conventional pipelines for ChIP-seq. ChIP-seq analysis focuses on calling “peaks” to identify where a protein of interest binds DNA, while cfChIP-seq often requires quantification of low-abundance signals from cancer chromatin. Additionally, input cell-free chromatin amounts are generally small and variable, and conventional indicators of data quality like peak number are not optimized for cfChIP-seq and cfMeDIP-seq. To address these challenges, we developed SNAP—a streamlined, containerized, Nextflow pipeline optimized for cell-free histone modification and DNA methylation profiling. Methods: We implemented key steps in cfChIP-seq and cfMeDIP-seq data processing using Nextflow and Docker containerization to create a reproducible and portable workflow. Through reanalysis of > 800 datasets spanning H3K4me3, H3K27ac, and DNA methylation, we established quality metrics that identify datasets with acceptable quality for downstream analyses. Optimum thresholds were established for the total number of DNA fragments and enrichment of fragments mapping to “on-target” versus “off-target” regions. We tested strategies for signal normalization to optimize comparability between samples. SNAP performs quality control assessment and normalization, as well as genotype fingerprinting to rule out sample swaps and confirm that epigenomic profiles derived from the same individual where expected. Results: Fragment count cut-offs were set at 2, 100, 000 for H3K4me3 and H3K27ac ChIP-seq, and 7, 700, 000 for MeDIP-seq. Enrichment score thresholds were established at 10.9, 10.3, and 3.2 for H3K4me3, H3K27ac, and MeDIP-seq, respectively. Filtering our cohort based on these thresholds maximized the ability to distinguish cancer vs. healthy plasma based on the signal at tumor-type-specific regulatory elements. Sample clustering based on SNP fingerprints reliably grouped samples derived from the same individual. SNAP has been successfully deployed on HPC clusters and is compatible with cloud computing platforms, ensuring accessibility for diverse research environments. Conclusion: We developed SNAP, a Nextflow-based pipeline specifically optimized for plasma cell-free chromatin epigenomic profiling, including H3K4me3 ChIP-seq, H3K27ac ChIP-seq, and MeDIP-seq. Customized quality control metrics enhanced the detection of tumor-associated epigenomic signatures in cancer plasma samples compared to healthy controls. SNAP provides a robust, reproducible, and portable solution for high-sensitivity plasma epigenomic profiling. Citation Format: Ze Zhang, Paulo Cordeiro, Brad Fortunato, Karl Semaan, Razane El Hajj Chehade, Gunsagar Gulati, Garyoung Lee, Shahabeddin Sotudian, Ji-Heui Seo, Matthew Freedman, Sylvan Baca. SNAP: Streamlined Nextflow Analysis Pipeline for profiling circulating histone modifications identifies tumor epigenomic signatures in cancer plasma [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 2504.