Presenting the most comprehensive, cross-tissue cell atlases to date: New findings from the Human Cell Atlas consortium

Researchers with the international Human Cell Atlas (HCA) consortium – whose goal is to map every cell type in the human body, but who until now have mostly focused on studying cells in individual organs and tissues or small tissue subsets – report a major feat: creation of detailed maps of more than a million individual cells across 33 organs, representing the most comprehensive, cross-tissue cell atlases to date. Openly available in four studies, the resulting data has many therapeutic implications, including informing understanding of common and rare diseases, vaccine development, anti-tumor immunology and regenerative medicine.  

In the first study, the Tabula Sapiens Consortium delivers a particularly broad cell atlas that provides a molecular definition of more than 400 cell types across 24 organs. They call this reference the “Tabula Sapiens” dataset. To assemble it, the authors used single-cell RNA sequencing (scRNA-seq) on nearly 500,000 live cells, including epithelial, endothelial, stromal, and immune cells, collected from multiple tissues from individual donors. The ability to analyze several tissues from single donors allowed for cross-tissue comparisons controlled for genetic background, age, environmental exposure, and epigenetic effects. Using the Tabula Sapiens dataset, the authors discovered several new insights into human cellular biology, including how the same gene can be spliced differently into different cell types and how clones of immune cells can be shared across tissues.

Single-cell atlases hold promise to help map the specific cell types in which disease genes act across the human body. Building them requires all cell types to be profiled – including cells that are hard to capture using scRNA-seq – and cells from larger numbers of individuals, meaning scientists need to collect and freeze tissues before analysis. In a second study in this package, Gökcen Eraslan et al. optimized single-nucleus RNA sequencing (snRNA-seq) to overcome the challenge of using frozen cells. They applied this technique to frozen samples from 8 healthy human organs from 16 donors and compiled a cross-tissue atlas of more than 200,000 nuclei profiles. Eraslan et al. used machine learning to associate cells in the atlas with thousands of single-gene diseases and complex genetic diseases and traits to discover cell types and gene programs that could be involved in disease.

Historically, scientists’ understanding of the human immune system has been largely constrained to the role of cells circulating in blood, but immune cells within tissues are critical to maintaining health. To better understand the function of immune cells across tissues, studies by Cecilia Dominguez Conde and colleagues and Chenqu Suo and colleagues profiled adult and developing immune cells, respectively. Using scRNA-seq, Dominquez Conde et al. surveyed innate and adaptive immune cells from 16 tissues across 12 adult donors, resulting in gene expression profiles for more than 300,000 cells. They developed a machine learning tool called “CellTypist” to aid in cell type annotation. The approach allowed the authors to identify nearly 101 immune cell types or states from more than one million cells, including previously underappreciated cell states. In a final study, Suo et al. used scRNA-seq, antigen-receptor sequencing and spatial transcriptomics of 9 prenatal tissues to generate a single-cell and spatial atlas of the developing immune system across gestation stages. The findings, which move beyond those from studies focused on one or a few organs in the developing immune system, reveal that blood and immune cell development occurs across many peripheral tissues, not just in primary hemopoietic organs.

The four studies and their findings are discussed further in a related Perspective by Zedao Liu and Zemin Zhang. “Collectively, these pan-tissue studies bring us closer to building a comprehensive human single-cell atlas,” they say.

***A related embargoed news briefing was held at 11:00 a.m. U.S. ET on 10 May, as a Zoom Webinar. Recordings of the briefing can be found here.***