The Human Cell Atlas: making 'cell space' for disease

A single change in DNA, RNA, proteins or cellular images can be useful as a biomarker of disease onset or progression. With high-throughput molecular phenotyping of single cells, it is now conceivable that the molecular changes occurring across thousands, or tens of thousands, of individual cells could additionally be considered as a disease biomarker. Transition to a disease state would then be reflected by the shifts in cell numbers and locations across a multidimensional space that is defined by the molecular content of cells. Realising this ambition requires a robust formulation of such a multidimensional 'cell space'. This is one of the goals of the recently launched Human Cell Atlas project. A second goal is to populate this 'cell space' with all cell types in the human body. Here, I consider the potential of the Human Cell Atlas project for improving our description and understanding of the cell-type specificity of disease.