arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.13346q-bio.GNcs.AIcs.CV
收藏

CellMaster: 单细胞分析中的协作细胞类型注释

CellMaster: Collaborative Cell Type Annotation in Single-Cell Analysis

Halicioglu Data Science Institute, University of California, San Diego, CA, USA(哈利奇奥格卢数据科学研究所,加州大学圣地亚哥分校);Department of Electrical & Computer Engineering, Texas A&M University, College Station, TX, USA(电气与计算机工程系,德克萨斯A&M大学);Department of Medicine, University of California, San Diego, CA, USA(医学系,加州大学圣地亚哥分校);Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, CA, USA(化学与生物化学系,加州大学圣地亚哥分校);Moores Cancer Center, University of California, San Diego, La Jolla, CA, USA(摩尔癌症中心,加州大学圣地亚哥分校);Mohamed bin Zayed University of AI, Abu Dhabi, UAE(穆罕默德·本·扎耶德人工智能大学);School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, USA(计算机科学学院,卡内基梅隆大学)

Zhen Wang, Yiming Gao, Jieyuan Liu, Enze Ma, Jefferson Chen, Mark Antkowiak, Mengzhou Hu, JungHo Kong, Dexter Pratt, Zhiting Hu, Wei Wang, Trey Ideker, Eric P. Xing

AI总结:

CellMaster通过利用LLM编码知识实现零样本细胞类型注释,提升单细胞分析的准确性和可解释性。

摘要:

单细胞RNA测序(scRNA-seq)能够对复杂组织进行大规模谱系分析,揭示罕见谱系和瞬时状态。然而,赋予生物学上有效的细胞身份仍是一个瓶颈,因为标记是组织和状态依赖性的,而新状态缺乏参考。我们提出了CellMaster,一个模仿专家实践的AI代理,用于零样本细胞类型注释。与现有自动化工具不同,CellMaster利用LLM编码的知识(例如GPT-4o)进行即时注释,具有可解释的推理理由,无需预训练或固定标记数据库。在覆盖8种组织的9个数据集中,CellMaster在自动模式下将准确性提高了7.1%(包括CellTypist和scTab在内的最佳基线模型)。在人机协作的精炼过程中,这一优势增加到18.6%,在亚型群体上获得22.1%的提升。该系统在罕见和新细胞状态中表现出特别的优势,其中基线模型往往失败。源代码和网络应用可在https://github.com/AnonymousGym/CellMaster获取。

英文摘要:

Single-cell RNA-seq (scRNA-seq) enables atlas-scale profiling of complex tissues, revealing rare lineages and transient states. Yet, assigning biologically valid cell identities remains a bottleneck because markers are tissue- and state-dependent, and novel states lack references. We present CellMaster, an AI agent that mimics expert practice for zero-shot cell-type annotation. Unlike existing automated tools, CellMaster leverages LLM-encoded knowledge (e.g., GPT-4o) to perform on-the-fly annotation with interpretable rationales, without pre-training or fixed marker databases. Across 9 datasets spanning 8 tissues, CellMaster improved accuracy by 7.1% over best-performing baselines (including CellTypist and scTab) in automatic mode. With human-in-the-loop refinement, this advantage increased to 18.6%, with a 22.1% gain on subtype populations. The system demonstrates particular strength in rare and novel cell states where baselines often fail. Source code and the web application are available at \href{https://github.com/AnonymousGym/CellMaster}{https://github.com/AnonymousGym/CellMaster}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑