美国密歇根大学Xiang Zhou团队近期取得重要工作进展，他们研究提出了准确高效的空间转录组学综合参考信息空间域检测计算方法。相关研究成果2024年6月6日在线发表于《自然—方法学》杂志上。

据介绍，空间分辨率转录组学（SRT）研究越来越普遍和广泛，为绘制复杂的组织结构和功能提供了前所未有的机会。

研究人员介绍了一种计算方法，综合参考组织分割法（IRIS），该方法旨在通过准确、高效地检测空间域来描述SRT研究中的组织空间组织特征。IRIS独特地利用单细胞RNA测序数据对生物可解释的空间结构域进行参考知情检测，整合多个SRT切片，同时明确考虑切片内和切片间的相关性。

研究人员通过对六个SRT数据集的深入分析，展示了IRIS的优势，这些数据集涵盖了不同的技术、组织、物种和分辨率。在这些应用中，IRIS在中等大小的数据集中实现了显著的准确性提高（39–1083%）和速度提高（4.6–666.0），同时是唯一适用于包括Stereo-seq和10x Xenium在内的大型数据集的方法。

因此，IRIS可以揭示复杂的大脑结构，发现肿瘤微环境的异质性，检测受糖尿病影响的睾丸的结构变化，而且速度和准确性都非常高。

附：英文原文

Title: Accurate and efficient integrative reference-informed spatial domain detection for spatial transcriptomics

Author: Ma, Ying, Zhou, Xiang

Issue&Volume: 2024-06-06

Abstract: Spatially resolved transcriptomics (SRT) studies are becoming increasingly common and large, offering unprecedented opportunities in mapping complex tissue structures and functions. Here we present integrative and reference-informed tissue segmentation (IRIS), a computational method designed to characterize tissue spatial organization in SRT studies through accurately and efficiently detecting spatial domains. IRIS uniquely leverages single-cell RNA sequencing data for reference-informed detection of biologically interpretable spatial domains, integrating multiple SRT slices while explicitly considering correlations both within and across slices. We demonstrate the advantages of IRIS through in-depth analysis of six SRT datasets encompassing diverse technologies, tissues, species and resolutions. In these applications, IRIS achieves substantial accuracy gains (39–1,083%) and speed improvements (4.6–666.0) in moderate-sized datasets, while representing the only method applicable for large datasets including Stereo-seq and 10x Xenium. As a result, IRIS reveals intricate brain structures, uncovers tumor microenvironment heterogeneity and detects structural changes in diabetes-affected testis, all with exceptional speed and accuracy.

DOI: 10.1038/s41592-024-02284-9

Source: https://www.nature.com/articles/s41592-024-02284-9