Towards semantic representations of tissue organization from high-parameter imaging data
Eric and Wendy Schmit Center Postdoctoral Fellow, Caicedo and Uhler labs, Broad Institute
Advances in imaging technologies present the opportunity to observe tissues at a high spatial and molecular resolution. Leveraging such data to construct an understanding of tissue organization requires building models for tissue semantics: conceptually and computationally defining the possible basic units of tissues and the rules governing their assembly and collective functionality. We will discuss four interrelated, but distinct, models for tissue semantics, inspired respectively by: urban neighborhoods, mechanical schematics, formal languages and genetic parts. I will highlight the complementary insights on tissue organization (both specific biology and general principles) offered by each of these models when applied to high-parameter imaging data, as well as the algorithmic opportunities they raise for increasing the depth and scale, at which tissues can be understood.