In the past decade, mass cytometry has been revolutionizing cytomics due to its ability for high‐dimensional characterization of single cells using isotopetagged antibodies enabling the simultaneous interrogation of 40+ protein targets. It thereby captures the complexity of cellular systems at unprecedented depth and has been key to advances in human systems biology, in particular in the field of immunology and cancer biology. Cytomics by mass cytometry has also become an integral part of multi‐OMICS studies. Recently, imaging mass cytometry and related techniques have been developed. These permit studying high‐dimensional cellular features in histological sections using a similar approach, thereby allowing the analysis of cellular phenotypes and their location in solid tissue in outstanding detail 1. Improvements in mass cytometry protocols have overcome many initial shortcomings and now allow reliable and standardized sample processing. Along with that, mass cytometry has transitioned into an established technology. However, pioneering sites initially faced struggles with implementing the highest standards of the technology. As a consequence, in 2016, a group of mass cytometry experts teamed up to connect and found the German Mass Cytometry Network (GerMaNet) to promote the further development, implementation, and applications of mass cytometry primarily in biomedical research. Today, the network spans platforms at the DRFZ Berlin, the Berlin Institute of Health (BIH), the Center for Regenerative Therapies in Dresden (CRTD), the TranslaTUM at the Technical University Munich, the Max‐Planck‐Institute for Molecular Genetics in Berlin (MPI‐MG), the University of Ulm, the Center for Molecular Medicine in Cologne, the Jena University Hospital, and the University Medical Center of Freiburg, with imaging units available at the BIH, MPI‐MG, and in Freiburg. Close interactions exist with platforms in Prague (Czech Republic) and other platforms in Europe, the United States, and Australia.

Mass cytometry projects are commonly highly multidisciplinary research efforts, and their progress is much quicker when domain experts collaborate to contribute their expertise to achieve highest quality standards for mass cytometry data generation and interpretation. Relevant areas include

• instrumentation and basic mass spectrometry,
• chemistry underlying assays and reagents,
• experimental workflows suitable for large scale trials and minimal batch effects,
• efficient data logistics and data preprocessing, and
• data analysis software tools suitable for handling of high‐dimensional single‐cell data.

The complexity of mass cytometry projects frequently presents significant risks for individual researchers. To minimize hurdles, the German Mass Cytometry Network introduced several strategies to facilitate the exchange of expertise across different sites and helped to critically advance several projects. These strategies included mutual visits, joint troubleshooting, assay and data acquisition training, and shadowing for specific protocols. GerMaNet also helps the swift distribution of new reagents and data analysis tools among the centers. This approach has proven highly effective, as reflected by several joint publications within the GerMaNet 2-6.

As an important integral part of overall networking efforts, GerMaNet started hosting small but impactful and well received annual meetings primarily for national audience, the German Mass Cytometry User Forum (Table 1). Steered by academic mass cytometry labs and core facilities, the Forum provides a hub for exchange in the mass cytometry community in central Europe.

The past three meetings in 2018, 2019, and 2020 have attracted primarily academicians from Germany, but also numerous international guests and participants from 14 European countries, Israel, and the United States. In addition to presentations by leading researchers in the field, young scientists and students were particularly addressed by integrating a poster session and selection of oral presentations from submitted abstracts. The gathering also provided a forum to discuss novel solutions regarding mass cytometry instrumentation, lab devices, reagents, assays, and data analysis, including commercial products.

As the mass cytometry community grew larger and the technology gained maturity through multiple iterative assay and hardware improvements, the focus of the Forum shifted from technical aspects toward customized workflows to address diverse applications in immunology, immune‐oncology, oncology, biology, nanotoxicology, biomarker identification, pathogenesis of immune‐mediated diseases such as rheumatoid arthritis, SLE and multiple sclerosis, spanning works with human, mouse and drosophila cells. This development parallels the results from mass cytometry workshops held at CYTO 7, 8. Many projects presented at the Forum were later published in peer‐reviewed journals 3, 4, 9-13.

The 3rd German Mass Cytometry User Forum took place from January 23‐24th in Berlin, Germany (Fig. 1). Speakers included Michael Leipold from Stanford University, giving this year`s ISAC lecture addressing the challenges of large mass cytometry studies that have been identified as a major hurdle for the implementation of mass cytometry to monitor clinical trials 7. There, unwanted data variation can be minimized by, for example, barcoding, and the careful decision of which samples to combine into a barcoded pool, implementation of instrument and assay controls, and antibody cocktail preservation to minimize reagent variation. Along with that, he stressed the importance of annotating and publicly sharing mass cytometry datasets, for example, via Mendeley, Cytobank, Immport, or FlowRepository.

Burkhard Becher (University of Zurich, Switzerland), Henrik Mei (DRFZ Berlin, Germany), and Marie‐Laure Yaspo (MPI for Molecular Genetics, Berlin, Germany) showcased different application areas of mass cytometry for patient immune profiling, exploring the phenotypical setup of a specialized cell type, and precision medicine. For example, immune profiling of Multiple Sclerosis patients revealed an expansion of T helper cells expressing CXCR4 and GM‐CSF in the blood, which may serve as a future therapeutic target in MS. Further, it was now discovered that human antibody‐secreting plasma cells forming the basis of humoral immunity and memory are composed of a variety of different phenotypes, potentially permitting differential regulation of PC subsets in their bone marrow environment. Finally, the integration of multiplexed pathology data from imaging mass cytometry along with genomic and transcriptional data in a multi‐OMICS approach was suggested. This promises direct impact on the care of cancer patients, to increase the benefit of precision medicine.

Mass cytometry hubs often maintain several collaborations, entailing the need for flexible and customizable data analysis solutions. In this regard, two examples for such pipelines were presented by Antonio Cosma (National Cytometry Platform, Luxembourg) and Thomas Höllt (Leiden University, The Netherlands), introducing a Tableau‐based workflow and Cytosplore (www.cytosplore.org), respectively. Tyler Burns (Berlin, Germany) then discussed kNN‐based preservation of local and global data structure preservation after dimensionality reduction by PCA, t‐SNE, and UMAP, with important implications for the reliability of gating and clustering in dimensionality‐reduced data space.

Two workshops were dedicated to news and burning questions in mass cytometry. In the Basics and Reagents workshop Michael Leipold, Antonio Cosma, Marjolijn Hameetman (LUMC, Leiden), Henrik Mei, and Axel Schulz (moderator) reviewed and discussed the current needs in the field, that is, to achieve technically consistent data across a large number of samples and measurements, expanding the measurement capacity of mass cytometry, and, to ease the setup of mass cytometry assays. Concerning the latter, benefits of novel pre‐made antibody panels suitable to characterize the most common immune cell populations 14 and designed for easy assay handling were discussed. The easy access to such commercially available panels can help speed up standard immune profiling by mass cytometry especially when combined with proprietary software for analyzing this assay: While this strategy may pave the way toward using mass cytometry in clinical practice, its customization is limited.

As doublets may present as artifacts in mass cytometry data, efforts aiming at their selective removal are ongoing. A novel workflow uses doublet discrimination based on Gaussian parameters 15. The panel agreed that this approach helps to improve doublet removal in the absence of barcoding but will need broader verification by the community. Doublet‐filtering sample barcoding 16, 17 and a moderate cell acquisition speed of approximately 300 cells per second appear as the currently most reliable option to minimize doublets. The discussion also centered around expanding the range of usable mass channels. Recent advances include antibody labeling with cadmium isotopes, adding up to seven channels for antibody‐based barcoding or additional analytes. Further, an amine‐reactive, column‐free metal labeling approach has become available, which suggests itself for very limited probe amounts, or reduction‐sensitive probes. The longer term stability of the conjugates and compatibility with a wider range of antibodies and other probes remains to be addressed. The panel highlighted the importance of in‐house antibody cocktail stabilization by cryopreservation for the field 10, a method that helps reducing unwanted data variation and pipetting errors. The panel further discussed the implementation of anchor controls for batch normalization and pointed out that the control samples should be as similar to the assay samples as possible. Here, implementing lyophilized PBMC pre‐labeled with tantalum were discussed as a possible solution, while important limitations arise from the fact that only those markers preserved in lyophilized cells can later be used for proper batch normalization. Finally, the panel encouraged industry to develop more advanced tools for assay standardization in mass cytometry, and to update standard reagents such as tuning solution and bead preparations to cover the expanding range of additional isotope masses now routinely measured at the far ends of the mass cytometers` detection range, such as yttrium and bismuth. Beads with gradually increasing metal content, similar as “Rainbow” calibration particles in flow cytometry, to calibrate and monitor the sensitivity mass cytometers more precisely were also desired. Recently described osmium‐labeled polystyrene beads 9 could serve as a platform for such developments. Finally, the panel agreed that a broader availability and use of live‐cell barcoding options, for example, using antibodies targeting CD45 or beta‐2‐microglobulin 2, 18, 19, could significantly help to increase data quality.

The Data Analysis workshop moderated by Marie Urbicht, featured Burkhard Becher, Tyler Burns, Thomas Höllt, and Lars Rønn Olsen (Technical University of Denmark). Paralleling the maturation of the mass cytometry field as a whole, this year's lively discussion veered off questions on specific algorithmic tools and more toward the challenges of analyzing larger mass cytometry studies such as batch normalization. A key topic was the challenge to how to distinguish technical from biological variation and how to reliably detect and deal with batch effects. The panel recommended identifying potential batch effects in a study by appropriate visualization, for example, by generating a dimensionality‐reduced plot colored by processing day of the samples. Normalization strategies based on anchor samples, which were published in 2019 12, 20 provide promising tools, however, they have yet to be systematically interrogated. Users were encouraged to exclude low‐quality samples or entire faulty runs from further analysis rather than risking the accuracy of analysis of the entire dataset.

Again, the panelists emphasized the importance of publishing mass cytometry datasets along with publications. Not only because this should be considered good scientific practice as, for example, defined by the MIFlowCyt guidelines 21, but also because it would allow for a more robust benchmarking of computational tools on more datasets, also comprising non‐hematopoietic cell types, such as solid tumors or tissue‐resident cells. Furthermore, the development of community‐wide standards on data annotation and analysis workflow documentation was identified as another mostly unmet need to ensure quality and reproducibility in dealing with highly multiplexed cytometry data.

The poster tour at the networking evening was organized by Désirée Kunkel (Berlin) and Sarah Warth (Ulm). Tomer Meir Salame (Weizmann Institute, Rehovot, Israel) was awarded this year`s poster prize for his presentation on the role of PD‐1/PD‐L1 and CCR2 in mouse models of Alzheimer`s Disease. Sufficient time was dedicated to personal communications to aid establishment new collaborations, importantly contributing to the success and serving the aims of the meeting and the network.

In sum, the German Mass Cytometry User Forum meetings have quickly established themselves as a platform for the national and international exchange of mass cytometry related research questions. The meetings have attracted growing numbers of attendees and industry, and increasingly received international attention (Table 1), demonstrating the demand for small and agile conferences that offer excellent opportunities for intimate scientific exchange regarding novel and specialized technologies in a supportive environment.

在过去的十年中，质谱流式细胞术已经彻底改变了细胞组学，因为它能够使用同位素标记的抗体对单个细胞进行高维表征，从而能够同时检测 40 多种蛋白质靶标。因此，它以前所未有的深度捕捉了细胞系统的复杂性，并且一直是人类系统生物学进步的关键，特别是在免疫学和癌症生物学领域。质谱流式细胞术的细胞组学也已成为多 OMICS 研究的一个组成部分。最近，已经开发了成像质量流式细胞术和相关技术。这些允许使用类似的方法研究组织切片中的高维细胞特征，从而可以非常详细地分析细胞表型及其在实体组织中的位置1. 质谱流式细胞术协议的改进克服了许多最初的缺点，现在允许可靠和标准化的样本处理。与此同时，质谱流式细胞术已经转变为一种成熟的技术。然而，开创性的网站最初面临着实施最高技术标准的困难。因此，2016年，一群质谱专家联手建立了德国质谱网络（GerMaNet），以推动质谱技术在生物医学研究领域的进一步发展、实施和应用。今天，该网络跨越了柏林 DRFZ、柏林卫生研究所 (BIH)、德累斯顿再生治疗中心 (CRTD)、慕尼黑工业大学 TranslaTUM、柏林马克斯普朗克分子遗传学研究所的平台(MPI-MG), 乌尔姆大学、科隆分子医学中心、耶拿大学医院和弗莱堡大学医学中心，在 BIH、MPI-MG 和弗莱堡提供成像装置。与布拉格（捷克共和国）的平台以及欧洲、美国和澳大利亚的其他平台存在密切互动。

质谱仪项目通常是高度多学科的研究工作，当领域专家合作贡献他们的专业知识以实现质谱仪数据生成和解释的最高质量标准时，它们的进展会更快。相关领域包括

• 仪器和基础质谱，
• 化学基础分析和试剂，
• 适用于大规模试验和最小批量效应的实验工作流程，
• 高效的数据物流和数据预处理，以及
• 适用于处理高维单细胞数据的数据分析软件工具。

大规模流式细胞术项目的复杂性经常给个别研究人员带来重大风险。为了最大程度地减少障碍，德国大规模流式细胞术网络引入了多种策略来促进不同站点之间的专业知识交流，并帮助关键地推进了几个项目。这些策略包括相互访问、联合故障排除、分析和数据采集培训以及特定协议的影子。GerMaNet 还有助于在中心之间快速分发新试剂和数据分析工具。正如 GerMaNet 2-6内的一些联合出版物所反映的那样，这种方法已被证明非常有效。

作为整体网络工作的重要组成部分，GerMaNet 开始举办小型但有影响力且广受好评的年度会议，主要面向全国观众，即德国质谱用户论坛（表 1）。该论坛由学术性的大规模流式细胞术实验室和核心设施指导，为中欧大规模流式细胞术社区的交流提供了一个枢纽。

2018年、2019年和2020年的三届会议主要吸引了来自德国的院士，但也有来自14个欧洲国家、以色列和美国的众多国际嘉宾和参与者。除了该领域领先研究人员的演讲外，还特别针对年轻科学家和学生进行了海报展示和从提交的摘要中选择口头演讲的活动。此次聚会还提供了一个论坛，讨论有关质谱仪、实验室设备、试剂、测定和数据分析（包括商业产品）的新解决方案。

随着大规模流式细胞术社区的发展壮大以及技术通过多次迭代分析和硬件改进逐渐成熟，论坛的重点从技术方面转向定制工作流程，以解决免疫学、免疫肿瘤学、肿瘤学、生物学、纳米毒理学、生物标志物的各种应用免疫介导的疾病（如类风湿性关节炎、SLE 和多发性硬化症）的鉴定、发病机制，跨越人类、小鼠和果蝇细胞的工作。这一发展与在 CYTO 7, 8举行的大规模流式细胞术研讨会的结果相似。论坛上展示的许多项目后来都发表在同行评审期刊3, 4, 9-13 上。

第三届德国质谱用户论坛于 1 月 23 日至 24 日在德国柏林举行（图 1）。演讲者包括来自斯坦福大学的 Michael Leipold，他在今年的 ISAC 演讲中探讨了大规模流式细胞术研究的挑战，这些挑战已被确定为实施大规模流式细胞术监测临床试验的主要障碍7。在那里，可以通过例如条形码、仔细决定将哪些样本组合到条形码池中、实施仪器和检测控制以及抗体鸡尾酒保存以最大限度地减少试剂变化来最大限度地减少不需要的数据变化。与此同时，他强调了注释和公开共享大规模细胞计数数据集的重要性，例如，通过 Mendeley、Cytobank、Immport 或 FlowRepository。

Burkhard Becher（瑞士苏黎世大学）、Henrik Mei（德国柏林 DRFZ）和 Marie-Laure Yaspo（分子遗传学 MPI，德国柏林）展示了质谱流式细胞术在患者免疫分析中的不同应用领域，探索了表型设置专门的细胞类型和精准医学。例如，多发性硬化症患者的免疫分析显示血液中表达 CXCR4 和 GM-CSF 的 T 辅助细胞扩增，这可能成为 MS 未来的治疗靶点。此外，现在发现形成体液免疫和记忆基础的人类抗体分泌浆细胞由多种不同的表型组成，可能允许对其骨髓环境中的 PC 亚群进行差异调节。最后，建议在多 OMICS 方法中整合来自成像质谱流式细胞术的多重病理数据以及基因组和转录数据。这有望直接影响癌症患者的护理，增加精准医疗的益处。

质谱流式细胞仪中心通常保持多个协作，因此需要灵活且可定制的数据分析解决方案。在这方面，Antonio Cosma（卢森堡国家细胞计量学平台）和 Thomas Höllt（荷兰莱顿大学）分别介绍了基于 Tableau 的工作流程和 Cytosplore（www.cytosplore.org）的两个此类管道示例。Tyler Burns（德国柏林）随后讨论了在通过 PCA、t-SNE 和 UMAP 降维后基于 kNN 的局部和全局数据结构保留，这对降维数据空间中门控和聚类的可靠性具有重要意义。

两个研讨会专门讨论质谱流式细胞术中的新闻和热点问题。在基础知识和试剂研讨会上，Michael Leipold、Antonio Cosma、Marjolijn Hameetman（LUMC，Leiden）、Henrik Mei 和 Axel Schulz（主持人）回顾并讨论了该领域的当前需求，即在大范围内实现技术上一致的数据。样品和测量的数量，扩大了质谱流式细胞仪的测量能力，并简化了质谱流式细胞仪检测的设置。关于后者，适用于表征最常见免疫细胞群的新型预制抗体组的好处14并讨论了为便于测定处理而设计的。轻松获取此类市售面板有助于通过质谱流式细胞术加速标准免疫分析，尤其是在与专有软件结合用于分析此检测时：虽然这种策略可能为在临床实践中使用质谱流式细胞术铺平道路，但其定制是有限的。

由于双峰可能作为大规模细胞计数数据中的伪影出现，因此正在进行旨在选择性去除双峰的努力。一种新颖的工作流程使用基于高斯参数15 的双重歧视。专家组一致认为，这种方法有助于在没有条形码的情况下改善双峰去除，但需要社区进行更广泛的验证。双峰过滤样本条码16, 17大约每秒 300 个细胞的中等细胞采集速度似乎是目前最可靠的选项，可以最大限度地减少双峰。讨论还围绕扩大可用大众渠道的范围展开。最近的进展包括用镉同位素标记抗体，为基于抗体的条形码或其他分析物添加多达七个通道。此外，胺反应性、无柱金属标记方法已经可用，这表明其适用于非常有限的探针量或还原敏感探针。缀合物的长期稳定性以及与更广泛的抗体和其他探针的兼容性仍有待解决。该小组强调了通过冷冻保存来稳定内部抗体混合物在现场的重要性10，一种有助于减少不需要的数据变化和移液错误的方法。该小组进一步讨论了批量标准化的锚控制的实施，并指出控制样本应尽可能类似于化验样本。在这里，实施用钽预标记的冻干 PBMC 作为一种可能的解决方案进行了讨论，而重要的限制来自这样一个事实，即只有保存在冻干细胞中的那些标记物以后才能用于适当的批次标准化。最后，该小组鼓励行业开发更先进的工具，用于质谱流式细胞术的测定标准化，并更新标准试剂，如调谐溶液和珠子制剂，以涵盖现在在质谱流式细胞仪远端常规测量的额外同位素质量的不断扩大范围`检测范围，如钇和铋。还需要金属含量逐渐增加的珠子，类似于流式细胞仪中的“彩虹”校准颗粒，以更精确地校准和监测灵敏度质量细胞仪。最近描述的锇标记的聚苯乙烯珠9可以作为此类发展的平台。最后，专家组一致认为，活细胞条形码选项的更广泛可用性和使用，例如，使用靶向 CD45 或 beta-2-微球蛋白2, 18, 19 的抗体，可以显着帮助提高数据质量。

该数据分析研讨会由 Marie Urbicht 主持，特色是 Burkhard Becher、Tyler Burns、Thomas Höllt 和 Lars Rønn Olsen（丹麦技术大学）。伴随着整个质谱流式细胞术领域的成熟，今年的热烈讨论转移了对特定算法工具的问题，更多地转向了分析更大的质谱流式细胞术研究（如批量归一化）的挑战。一个关键主题是如何区分技术变异和生物变异以及如何可靠地检测和处理批次效应的挑战。该小组建议通过适当的可视化来识别研究中潜在的批次效应，例如，通过生成按样品处理天数着色的降维图。基于anchor样本的归一化策略，发表于2019年12, 20提供有希望的工具，但是，它们尚未被系统地审问。鼓励用户从进一步分析中排除低质量样本或整个错误运行，而不是冒整个数据集分析的准确性的风险。

小组成员再次强调了与出版物一起发布大规模细胞计数数据集的重要性。不仅因为这应该被视为良好的科学实践，例如由 MIFlowCyt 指南21定义，而且因为它允许对更多数据集的计算工具进行更可靠的基准测试，还包括非造血细胞类型，例如实体瘤或组织驻留细胞。此外，关于数据注释和分析工作流程文档的社区标准的制定被确定为另一个主要未满足的需求，以确保处理高度多路复用的细胞计数数据的质量和可重复性。

联谊晚会的海报巡展由 Désirée Kunkel（柏林）和 Sarah Warth（乌尔姆）组织。Tomer Meir Salame（以色列雷霍沃特魏茨曼研究所）因其关于 PD-1/PD-L1 和 CCR2 在阿尔茨海默病小鼠模型中的作用的演讲而获得了今年的海报奖。有足够的时间用于个人交流，以帮助建立新的合作，重要的是有助于会议和网络的成功和服务的目标。

总之，德国大规模流式细胞术用户论坛会议已迅速成为国内和国际大规模流式细胞术相关研究问题交流的平台。这些会议吸引了越来越多的与会者和行业，并越来越受到国际关注（表 1），这表明对小型和敏捷会议的需求，这些会议为在支持性环境中就新颖和专业技术进行亲密的科学交流提供了绝佳机会。