肿瘤克隆的异质性驱动了不同肿瘤细胞群的选择和进化，导致复杂且动态的肿瘤进化过程。虽然肿瘤批量 DNA 测序有助于阐明肿瘤内异质性，但由于拷贝数变化和重建过程中的不确定性而导致的突变多重性错误识别等挑战阻碍了肿瘤进化的准确推断。在本研究中，我们引入了一种新方法，即重建肿瘤克隆异质性和进化关系（RETCHER），该方法通过准确识别突变多重性，同时考虑重建过程中的不确定性以及亚克隆聚类的可信度和合理性，来表征更真实的癌细胞组分。该方法全面、准确地推断了多种形式的肿瘤克隆异质性和系统发育关系。 RETCHER 在模拟数据上优于现有方法，并在五种肿瘤类型的真实多样本测序数据中推断出更清晰的亚克隆结构和进化关系。通过精确分析肿瘤内复杂的克隆异质性，RETCHER为肿瘤进化研究提供了新方法，并为制定精准、个性化的治疗策略提供了科学证据。该方法有望在肿瘤进化研究、临床诊断和治疗中发挥重要作用。 RETCHER 可在 https://github.com/zlsys3/RETCHER 上免费获取。© 作者 2024。由牛津大学出版社出版。

The heterogeneity of tumor clones drives the selection and evolution of distinct tumor cell populations, resulting in an intricate and dynamic tumor evolution process. While tumor bulk DNA sequencing helps elucidate intratumor heterogeneity, challenges such as the misidentification of mutation multiplicity due to copy number variations and uncertainties in the reconstruction process hinder the accurate inference of tumor evolution. In this study, we introduce a novel approach, REconstructing Tumor Clonal Heterogeneity and Evolutionary Relationships (RETCHER), which characterizes more realistic cancer cell fractions by accurately identifying mutation multiplicity while considering uncertainty during the reconstruction process and the credibility and reasonableness of subclone clustering. This method comprehensively and accurately infers multiple forms of tumor clonal heterogeneity and phylogenetic relationships. RETCHER outperforms existing methods on simulated data and infers clearer subclone structures and evolutionary relationships in real multisample sequencing data from five tumor types. By precisely analysing the complex clonal heterogeneity within tumors, RETCHER provides a new approach to tumor evolution research and offers scientific evidence for developing precise and personalized treatment strategies. This approach is expected to play a significant role in tumor evolution research, clinical diagnosis, and treatment. RETCHER is available for free at https://github.com/zlsys3/RETCHER.© The Author(s) 2024. Published by Oxford University Press.