通过亚硫酸氢盐处理的游离 DNA (cfDNA) 片段进行早期癌症诊断需要繁琐的数据分析程序。在这里，我们提出了一种基于深度学习的早期癌症拦截和诊断方法 (DECIDIA)，该方法可以仅通过亚硫酸氢盐处理的 cfDNA 测序片段实现准确的癌症诊断。 DECIDIA 依赖于基于 Transformer 的 DNA 片段表示学习和弱监督多实例学习来进行分类。我们在包含 5389 个样本的精选数据集上系统地评估了 DECIDIA 在癌症诊断和癌症类型预测方面的性能，这些样本包括结直肠癌 (CRC；n = 1574)、肝细胞细胞癌 (HCC；n = 1181)、肺癌 (n = 654）和非癌症对照（n = 1980）。通过区分癌症患者与无癌对照，DECIDIA 在 CRC 数据集的 10 倍交叉验证设置中实现了 0.980 的受试者工作曲线下面积 (AUROC)（95% CI，0.976-0.984），优于以下基准方法：基于甲基化强度。值得注意的是，尽管模型开发中没有使用 HCC 数据，但 DECIDIA 在外部独立 HCC 测试集上实现了 0.910 的 AUROC（95% CI，0.896-0.924），以区分 HCC 患者与无癌症对照。在癌症类型分类的设置中，我们观察到 DECIDIA 的微平均 AUROC 为 0.963（95% CI，0.960-0.966），总体准确度为 82.8%（95% CI，81.8-83.9）。此外，我们从原始测序读数中提取了四个序列特征，这些特征在癌症与对照以及不同癌症类型之间表现出差异模式。我们的方法代表了一种新的范式，旨在消除使用亚硫酸氢盐处理的 cfDNA 甲基组进行液体活检的繁琐数据分析程序。© 2024 作者。约翰·威利出版的《分子肿瘤学》

Early cancer diagnosis from bisulfite-treated cell-free DNA (cfDNA) fragments requires tedious data analytical procedures. Here, we present a deep-learning-based approach for early cancer interception and diagnosis (DECIDIA) that can achieve accurate cancer diagnosis exclusively from bisulfite-treated cfDNA sequencing fragments. DECIDIA relies on transformer-based representation learning of DNA fragments and weakly supervised multiple-instance learning for classification. We systematically evaluate the performance of DECIDIA for cancer diagnosis and cancer type prediction on a curated dataset of 5389 samples that consist of colorectal cancer (CRC; n = 1574), hepatocellular cell carcinoma (HCC; n = 1181), lung cancer (n = 654), and non-cancer control (n = 1980). DECIDIA achieved an area under the receiver operating curve (AUROC) of 0.980 (95% CI, 0.976-0.984) in 10-fold cross-validation settings on the CRC dataset by differentiating cancer patients from cancer-free controls, outperforming benchmarked methods that are based on methylation intensities. Noticeably, DECIDIA achieved an AUROC of 0.910 (95% CI, 0.896-0.924) on the externally independent HCC testing set in distinguishing HCC patients from cancer-free controls, although there was no HCC data used in model development. In the settings of cancer-type classification, we observed that DECIDIA achieved a micro-average AUROC of 0.963 (95% CI, 0.960-0.966) and an overall accuracy of 82.8% (95% CI, 81.8-83.9). In addition, we distilled four sequence signatures from the raw sequencing reads that exhibited differential patterns in cancer versus control and among different cancer types. Our approach represents a new paradigm towards eliminating the tedious data analytical procedures for liquid biopsy that uses bisulfite-treated cfDNA methylome.© 2024 The Author(s). Molecular Oncology published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.