Integrative interpretable learning reveals shared patterns of epitranscriptomic regulation across multiple cancer types

Cancer is a complex set of diseases caused by the dysregulation of cell proliferation, differentiation, and apoptosis, ultimately leading to malignant tumor development and metastasis. Recent studies have revealed that dysregulation of N4-acetylcytidine (ac4C) acetylation is associated with enhanced metastatic potential and tumor progression in various cancer types. However, it remains unclear whether diverse cancer types share common epitranscriptomic regulatory patterns or engage in interconnected networks. In this study, leveraging an extensive collection of 88 ac4C epitranscriptome datasets profiled across multiple cancer types and normal conditions, we developed the first ac4C pan-cancer model by integrating a diverse set of sequence and curated genome-derived knowledge, based on a combined deep learning-powered transformer architecture. Our interpretable analysis uncovered, for the first time, the shared epitranscriptomic patterns and associations of dysregulated ac4C across different cancers, particularly associated with low GC-content regions of the 3′UTR and internal 3′UTR splicing. Furthermore, we discovered a set of candidate ac4C-mediated genes that may function in cancers through epitranscriptomic regulation, including three novel ac4C-mediated candidates (SMARCD1, SENP5, RNF207) that exhibit consistently dysregulated ac4C levels, expression patterns, and are associated with poor clinical prognosis across different cancer types. Taken together, our findings highlight the importance of a comprehensive characterization of ac4C epitranscriptome in pan-cancer landscape, with potential implications for developing RNA modification-based biomarkers and therapeutic strategies.