Nazaret Achille, Fan Joy Linyue, Lavallée Vincent-Philippe, Burdziak Cassandra, Cornish Andrew E, Kiseliovas Vaidotas, Bowman Robert L, Masilionis Ignas, Chun Jaeyoung, Eisman Shira E, Wang James, Hong Justin, Shi Lingting, Levine Ross L, Mazutis Linas, Blei David, Pe'er Dana, Azizi Elham
Department of Computer Science, Columbia University, New York, NY 10027, USA.
Irving Institute for Cancer Dynamics, Columbia University, New York, NY 10027, USA.
bioRxiv. 2024 Nov 5:2023.11.11.566719. doi: 10.1101/2023.11.11.566719.
Biological insights often depend on comparing conditions such as disease and health, yet we lack effective computational tools for integrating single-cell genomics data across conditions or characterizing transitions from normal to deviant cell states. Here, we present Decipher, a deep generative model that characterizes derailed cell-state trajectories. Decipher jointly models and visualizes gene expression and cell state from normal and perturbed single-cell RNA-seq data, revealing shared and disrupted dynamics. We demonstrate its superior performance across diverse contexts, including in pancreatitis with oncogene mutation, acute myeloid leukemia, and gastric cancer.
生物学见解通常依赖于比较疾病与健康等状况，然而我们缺乏有效的计算工具来整合不同状况下的单细胞基因组学数据，或刻画从正常细胞状态到异常细胞状态的转变。在此，我们展示了Decipher，这是一种用于刻画偏离正轨的细胞状态轨迹的深度生成模型。Decipher联合建模并可视化来自正常和受干扰的单细胞RNA测序数据的基因表达和细胞状态，揭示共享和 disrupted 动态。我们在多种情况下证明了它的卓越性能，包括在伴有癌基因突变的胰腺炎、急性髓系白血病和胃癌中。 （注：原文中“disrupted”直译为“破坏的”，结合语境这里可能不太准确，但根据要求未作调整，整体翻译可能需要进一步根据专业知识优化表述。）
