Automated Profiling of Gene Function During Embryonic Development

Systematic functional profiling of the gene set that supports multicellular organismal development is an important current challenge. Here, we tackle this challenge in the context of C. elegans embryogenesis by capturing the evolution of gene knockdown phenotypes using 4D imaging and developing an  automated approach to analyze and compare complex, diverse developmental phenotypes. The automated approach quantifies embryonic features— including germ layer cell numbers, tissue position, and tissue shape—to generate temporal feature curves whose parametrization yields numerical  phenotypic signatures. In conjunction with a new similarity metric that operates across phenotypic space, these signatures enabled comparison of 500 gene knockdowns. Accessible in a redesigned Phenobank web-portal, this dataset represents ~25% of essential development genes. This approach correctly classifies known developmental regulators, annotates functions for uncharacterized genes, identifies new gene and pathway relationships, and establishes the foundation for comprehensive analysis of the gene  set required to build a multicellular organism.