In toto analysis of multicellular arrangement reduces embryonic tissue diversity to two archetypes that require specific cadherin expression

Breakthroughs in transcriptomics are providing exciting new ways to characterise cellular diversity within organisms1-3. Yet, organisms are more than the sum of their differentiated cells, as the biological function of most cell types only emerges when they are organised into tissues with characteristic architectures4. The mechanisms that drive architectural diversification at the tissue level remain poorly understood, in part due to a general lack of methods for directly comparing different organisational patterns. Here we establish nuQLOUD, an efficient imaging and computational framework that reduces complex tissues to clouds of nuclear positions, enabling the extraction of cell-type agnostic architectural features. Applying nuQLOUD to whole developing zebrafish embryos reveals that global tissue diversity can be efficiently reduced to two primary archetypes, termed amorphous and crystalline. We investigate the molecular drivers of tissue archetypes by focussing on cadherin cell adhesion molecules and demonstrate that the expression domains of major cadherins segregate along tissue-archetypal lines. Further spatiotemporal analysis and targeted perturbation of cadherin expression in different organs identifies N-Cadherin as a general driver of the amorphous archetype. Thus, this systematic investigation of architectural diversity provides a new way to conceptualise embryonic organisation and understand drivers of tissue diversification within a standardised framework.