Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Catching up with multiplexed tissue imaging

Highly multiplexed tissue imaging continues to show its power for biomedical discovery. In this issue, we publish tools and guidance for implementing this class of methods and reporting subsequent results.

Immunofluorescence imaging of tissues has been a staple in biomedical research, enabling researchers to gain rich information that can complement what can be seen using traditional approaches such as histological staining. Immunofluorescence imaging offers high spatial resolution over large regions and is uniquely suited to imaging the location and abundance of specific proteins of interest. Recent years have seen immunofluorescence pushed seemingly to its limits in terms of multiplexing, with well over 20 targets being readily accessible in a given experiment.

These highly multiplexed tissue imaging experiments (including CyCIF, mIHC, IMC, MELC, mxIF, CODEX and MIBI) can seem like an alphabet soup, but they tend to have methodological commonalities. Most use antibody-based staining to gain target specificity combined with either a fluorescence microscopy or imaging mass spectrometry readout. Many of the methods involve cyclic imaging or multiple rounds of staining to achieve high levels of multiplexing.

Although the proteome coverage of these methods does not yet approach what can be achieved for spatial transcriptomics methods, this level of multiplexing can already be used to distinguish among cell types and give detailed insight into tissue architecture, with single-cell resolution. Indeed, these techniques fill a methodological gap yet unfilled by single-cell proteomics approaches and are a crucial complement to conventional and spatial omics technologies targeting the genome and transcriptome. As such, several large consortia including the Human BioMolecular Atlas Program (HuBMAP), the Human Tumor Atlas Network (HTAN) and the LifeTime Initiative are working toward creating atlases combining data from highly multiplexed tissue imaging with single-cell omics data. Such atlases will yield unique insight into tissue structure and function in health and disease.

As with many areas of fast-paced methods growth, sometimes analytical tools, best practices guidelines and optimal reporting standards lag behind early waves of development and biological applications. Three papers published in this issue aim to address these open areas and improve multiplexed tissue imaging in practice.

A Perspective from Andrea Radtke, Sinem Saka and colleagues provides guidance for generating robust and reproducible imaging data using antibody-based multiplexing approaches. This piece integrates advice from domain experts and methods developers to help lower the barrier to new users getting optimal results with what can be complicated and nuanced protocols. In addition to describing numerous multiplexed imaging approaches, the authors outline strategies for preparing both samples and labeling reagents, provide guidance for rigorous antibody validation, and describe how to build suitable multiplexed antibody panels. They also offer an in-depth discussion of the unique aspects of processing, analyzing and storing highly multiplexed imaging data. Additionally, they provide an invaluable list of community-validated antibody clones against human and mouse proteins suitable for highly multiplexed imaging.

A Comment from Sandro Santagata, Peter Sorger and colleagues including the HTAN consortium puts forth the Minimum Information about Highly Multiplexed Tissue Imaging (MITI) standard. This work seeks to keep the reporting of these multiplexed imaging data at pace with standards developed for genomics and other microscopy data. The authors emphasize that the imminent release of atlas-type datasets combining highly multiplexed imaging and omics data reveals an immediate need for data and metadata standards that facilitate data curation and sharing. The piece promotes MITI guidelines and implementations, describes the governance and future of MITI, and notes that MITI consortium guidelines are designed to be broadly applicable and open to updates. Such guidelines are necessary to improve the reproducibility and reuse of such highly multiplexed data.

A Brief Communication from Peter Sorger and colleagues describing Multiple Choice MICROscopy (MCMICRO) rounds out the series. MCMICRO is a computational pipeline for analyzing whole-slide images of highly multiplexed tissue imaging data that meets the unique needs of these types of datasets and offers powerful analytical capabilities. MCMICRO is open access and allows for modular customization of its workflow, ultimately enabling researchers to handle all analytical steps from whole-slide images to single-cell data; it is also compatible with many types of highly multiplexed image data and overcomes issues associated with very large datasets. In the spirit of usability, MCMICRO is implemented in Nextflow and Galaxy, which offer a command line interface and graphical user interface, respectively. The software is extensively documented, and numerous resources are available to implement the many tools included in the pipeline.

Despite our longstanding interest in imaging and microscopy, Nature Methods does not have a long history of publishing methods for highly multiplexed tissue imaging. Two notable exceptions are our 2014 paper from Detlef Günther, Bernd Bodenmiller and colleagues describing the first application of CyTOF to imaging tumor tissues and the 2017 paper from Bodenmiller and colleagues describing histoCAT, a computational analysis toolbox for interactive and quantitative analysis of multiplexed imaging data.

Tissue imaging has long been the domain of histologists and pathologists, with an emphasis on diagnostic imaging. At Nature Methods, our editorial scope is fixed squarely on methods for basic research. This precludes diagnostic, preclinical, clinical and translational methods, including digital pathology — a methods space that has exploded, especially with the rise of deep learning for image analysis. That said, this area continues to grow and increase its reach into basic research, most notably in areas such as immunology and cancer biology, aligning it more closely with our editorial scope.

We expect that advances in probes, barcoding strategies, affinity reagents, multiplexed detection, microscope sensitivity and microscope throughput will continue to move highly multiplexed tissue imaging forward in important ways. We also think improved methods for integrating multiplexed imaging data with other omics technologies will be crucial to the next phases of technological advancement. Finally, improved algorithmic and software tools for viewing and analyzing highly multiplexed and multi-modal omics data will also be needed to make the most of these advances. We welcome submissions that push the limits of these technologies to enable new biological discovery.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Catching up with multiplexed tissue imaging. Nat Methods 19, 259 (2022). https://doi.org/10.1038/s41592-022-01428-z

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41592-022-01428-z

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing