Although scRNA-seq is now ubiquitously adopted in studies of intratumor heterogeneity, detection of somatic mutations and inference of clonal membership from scRNA-seq is currently unreliable. We propose DENDRO, an analysis method for scRNA-seq data that clusters single cells into genetically distinct subclones and reconstructs the phylogenetic tree relating the subclones. DENDRO utilizes transcribed point mutations and accounts for technical noise and expression stochasticity. We benchmark DENDRO and demonstrate its application on simulation data and real data from three cancer types. In particular, on a mouse melanoma model in response to immunotherapy, DENDRO delineates the role of neoantigens in treatment response.

中文翻译：

---

DENDRO：通过单细胞 RNA 测序进行遗传异质性分析和亚克隆检测


尽管 scRNA-seq 现在在肿瘤内异质性研究中普遍采用，但从 scRNA-seq 检测体细胞突变和推断克隆成员目前并不可靠。我们提出了 DENDRO，一种 scRNA-seq 数据的分析方法，它将单个细胞聚类成遗传上不同的亚克隆，并重建与亚克隆相关的系统发育树。 DENDRO 利用转录点突变并解释技术噪音和表达随机性。我们对 DENDRO 进行了基准测试，并展示了其在三种癌症类型的模拟数据和真实数据上的应用。特别是，在响应免疫治疗的小鼠黑色素瘤模型上，DENDRO 描述了新抗原在治疗反应中的作用。