Biomolecular condensates concentrate select groups of macromolecules into discrete foci in eukaryotic cells in the absence of a surrounding membrane. Condensates are found throughout eukaryotic cells, and function in processes ranging from signal transduction to RNA metabolism to gene expression. Aberrant condensates have been implicated in neurodegeneration, cancer and skin diseases. Our understanding of eukaryotic cell organization was transformed by the discovery that many condensates appear to form through liquid-liquid phase separation (LLPS) of multivalent macromolecules. My lab played an important part in this discovery by establishing key principles, including the essential role of multivalent interactions in promoting biological LLPS, the regulation of LLPS by covalent modifications, and the ability of LLPS to increase enzymatic activity. In recent years, we showed that and how LLPS can produce membrane-associated clusters that increase the specific activity of signaling molecules. We also showed that intrinsically disordered regions of proteins (IDRs) can undergo LLPS to produce liquid droplets that harden to solids over time, likely due to formation of amyloid filaments, and that misregulated hardening may contribute to neurodegeneration. Further, we proposed the first model to explain condensate composition based on a scaffold/client framework. Most recently, we showed that chromatin has an intrinsic propensity to undergo LLPS, providing a new view of eukaryotic genome organization. Here, we propose a broad program to address leading questions in the biomolecular condensate field. We will use microfluidics to assess LLPS of thousands of IDRs in a single experiment, leading to a predictive model for the sequence determinants of LLPS and amyloid formation by IDRs. This work will deepen our biophysical understanding of these processes, predict which IDRs are likely to contribute to specific biological processes (e.g. transcription) through self-assembly, and reveal how IDR mutations lead to disease through amyloid formation. We will also examine how individual components of yeast P bodies impact the LLPS threshold and composition of the compartments and how RNA helicase and RNA decapping activities are modulated within them. This work will lead to a new model of condensate composition based on the patterns of interaction between components, and will explain how composition and encapsulation can control enzymatic activities in native condensates. Finally, we will learn how internucleosome spacing and diverse histone post-translational modifications control chromatin LLPS to generate biochemically and functionally distinct genomic regions, examine whether NUT carcinoma is caused by defective LLPS, and develop a new approach to drug design based on targeting small molecules to condensates. Together, the work will reveal new principles of biological phase separation, explain how phase separation can be used to control RNA metabolism and geno...