Biological systems are functionally defined by the nature, amount and spatial location of the totality of their proteins. Resolving the spatiotemporal distribution of the human proteome at a subcellular level increases our understanding of human biology and disease. In the Human Protein Atlas project, we are systematically mapping the human proteome in a multitude of human cells and organs using microscopy. We have generated a high-resolution map of the subcellular distribution of the human proteome and have shown that as much as half of all proteins localize to multiple compartments. Such proteins may have context specific functions and ‘moonlight’ in different parts of the cell, thus increasing the functionality of the proteome and the complexity of the cell from a systems perspective.

Recently, we performed a single cell spatiotemporal dissection of the transcriptome and proteome over the cell cycle. We could identify 20% of the human proteome to display cell-to-cell variability, and present the first evidence of cell cycle association for 301 proteins. Our results show that cell cycle progression explains less than half of all cell-to-cell variability, and that most cycling proteins are regulated post-translationally, rather than by transcriptomic cycling.

All this work is critically dependent on computational image analysis, and I will discuss machine learning approaches for classification and embedding of spatial subcellular patterns, including the results from two recent Kaggle competitions.

In summary, I will demonstrate the importance of spatial proteomics data for improved single cell biology and present how the freely available Human Protein Atlas database (www.proteinatlas.org) can be used as a resource for life science.