Multimodal AI for Single cfDNA Profiling and Cancer Screening

Abstract Cell-free DNA (cfDNA) serves as a non-invasive biomarker for cancer detection, but conventional methods face challenges due to the ultra-low abundance of tumor-derived cfDNA (ctDNA) among normal cfDNA. Though nucleosome-bound cfDNA harbors rich epigenomic features that could enable ctDNA identification by single-molecule multi-omics cross-validation, this remains unexplored due to methodological limits. Here, we developed a cfDNA sequencing approach integrating methylation, fragmentomics, and histone modifications at the single-molecule level; together with gene semantics and epigenomic annotations, these modalities were vectorized and fused to represent each cfDNA molecule. We trained a Transformer-based model (cfAI) to profile and evaluate ctDNA likelihood at molecule, gene, and sample levels. cfAI achieved ∼10-fold enrichment of cancer-derived signals over noise and reached 72.6% sensitivity at 93.1% specificity for multi-cancer detection. Our study establishes an innovative framework that overcomes the inherent signal-to-noise limitations of conventional assays and reveals biological features at molecular resolution for cancer detection. Competing Interest Statement Liyang Song, Ying Xin, Guoqiang Zhao, Guo Chen, Jing Liu, Baoliang Zhu, Xueguang Sun and Xiaohui Wu are employees of Shanghai Xiaohe Medical Laboratory Co., Ltd, Shanghai, China. Liyang Song, Xueguang Sun and Xiaohui Wu are authors of the related patents. All other authors have declared no conflicts of interest.