School of Statistics and Mathematics, Zhongnan University of Economics and Law, Wuhan, 430073, China.
School of Information Engineering, Zhongnan University of Economics and Law, Wuhan, 430073, China.
Nat Commun. 2024 Sep 19;15(1):8223. doi: 10.1038/s41467-024-52445-9.
Detection and Dissection of Anomalous Tissue Domains (DDATD) from multi-sample spatial transcriptomics (ST) data provides unprecedented opportunities to characterize anomalous tissue domains (ATDs), revealing both population-level and individual-specific pathogenic factors for understanding pathogenic heterogeneities behind diseases. However, no current methods can perform de novo DDATD from ST data, especially in the multi-sample context. Here, we introduce STANDS, an innovative framework based on Generative Adversarial Networks which integrates three core tasks in multi-sample DDATD: detecting, aligning, and subtyping ATDs. STANDS incorporates multimodal-learning, transfer-learning, and style-transfer techniques to effectively address major challenges in multi-sample DDATD, including complications caused by unalignable ATDs, under-utilization of multimodal information, and scarcity of normal ST datasets necessary for comparative analysis. Extensive benchmarks from diverse datasets demonstrate STAND's superiority in identifying both common and individual-specific ATDs and further dissecting them into biologically distinct subdomains. STANDS also provides clues to developing ATDs visually indistinguishable from surrounding normal tissues.
从多样本空间转录组学 (ST) 数据中检测和剖析异常组织域 (DDATD) 为表征异常组织域 (ATD) 提供了前所未有的机会，揭示了理解疾病背后致病异质性的群体水平和个体特异性致病因素。然而，目前没有方法可以从头开始从 ST 数据中进行 DDATD，特别是在多样本的情况下。在这里，我们介绍了 STANDS，这是一个基于生成对抗网络的创新框架，它整合了多样本 DDATD 中的三个核心任务：检测、对齐和亚型化 ATD。STANDS 结合了多模态学习、迁移学习和风格迁移技术，有效地解决了多样本 DDATD 中的主要挑战，包括由不可对齐的 ATD 引起的复杂情况、多模态信息利用不足以及进行比较分析所需的正常 ST 数据集的稀缺。来自不同数据集的广泛基准测试表明，STANDS 在识别常见和个体特异性 ATD 方面具有优越性，并进一步将它们细分为具有生物学差异的子域。STANDS 还为开发与周围正常组织在视觉上无法区分的 ATD 提供了线索。
