The Human Cell Atlas

Abstract

The recent advent of methods for high-throughput single-cell molecular profiling has catalyzed a growing sense in the scientific community that the time is ripe to complete the 150-year-old effort to identify all cell types in the human body. The Human Cell Atlas Project is an international collaborative effort that aims to define all human cell types in terms of distinctive molecular profiles (such as gene expression profiles) and to connect this information with classical cellular descriptions (such as location and morphology). An open comprehensive reference map of the molecular state of cells in healthy human tissues would propel the systematic study of physiological states, developmental trajectories, regulatory circuitry and interactions of cells, and also provide a framework for understanding cellular dysregulation in human disease. Here we describe the idea, its potential utility, early proofs-of-concept, and some design considerations for the Human Cell Atlas, including a commitment to open data, code, and community.

Keywords: cell atlas; cell biology; computational biology; human; lineage; mouse; science forum; single-cell genomics; systems biology.

Figure 1.. A hierarchical view of human anatomy.

A graphical depiction of the anatomical hierarchy from organs (such as the gut), to tissues (such as the epithelium in the crypt in the small intestine), to their constituent cells (such as epithelial, immune, stromal and neural cells).

Figure 2.. Anatomy: cell types and tissue structure.

The first three plots show single cells (dots) embedded in low-dimensional space based on similarities between their RNA-expression profiles (A, C) or protein-expression profiles (B), using either t-stochastic neighborhood embedding (A,B) or circular projection (C) for dimensionality reduction and embedding. (A) Bi-polar neurons from the mouse retina. (B) Human bone marrow immune cells. (C) Immune cells from the mouse spleen. (D) Histology. Projection of single-cell data onto tissue structures: image shows the mapping of individual cells onto locations in the marine annelid brain, based on the correspondence (color bar) between their single-cell expression profiles and independent FISH assays for a set of landmark transcripts.

Figure 3.. Developmental trajectories.

Each plot shows single cells (dots; colored by trajectory assignment, sampled time point, or developmental stage) embedded in low-dimensional space based on their RNA (A-C) or protein (D) profiles, using different methods for dimensionality reduction and embedding: Gaussian process patent variable model (A); t-stochastic neighborhood embedding (B, D); diffusion maps (C). Computational methods then identify trajectories of pseudo-temporal progression in each case. (A) Myoblast differentiation in vitro. (B) Neurogenesis in the mouse brain dentate gyrus. (C) Embryonic stem cell differentiation in vitro. (D) Early hematopoiesis.

Figure 4.. Physiology.

Each plot shows single cells (dots) embedded in low-dimensional space on the basis of their RNA profile, based on predefined gene signatures (A) or PCA (B, C), highlighting distinct dynamic processes. (A) The cell cycle in mouse hematopoietic stem and progenitor cells; adapted under terms of CC BY 4.0 from Scialdone et al. (2015). (B) Response to lipopolysaccharide (LPS) in mouse immune dendritic cells. (C) Variation in the extent of pathogenicity in mouse Th17 cells.