Project 3/Summary NUP98 fusion oncogenes (e.g., NUP98-HOXA9, -KDM5A, and -NSD1) are associated with several hematological malignancies, including pediatric acute myeloid leukemia (AML), characterized by dismal treatment outcomes. The N-terminal FG-repeat domain that is common to all NUP98 fusion oncogenes (termed NUP98-N) is known to be intrinsically disordered and, by itself, to undergo phase separation to form gel-like bodies in vitro and large, spherical puncta in the nuclei of cells. However, these large puncta are not associated with hematopoietic cell transformation and leukemogenesis. Importantly, NUP98 fusion oncoproteins (FOs)—in which NUP98-N is fused to a variety of different C-terminal chromatin targeting domains—form many chromatin-associated, sub-micron-sized puncta that are associated with aberrant gene transcription, hematopoietic cell transformation and leukemogenesis. Recently, Young, et al., Tjian, et al., and Rivera, et al., demonstrated that critical transcriptional regulators, i) exhibit disordered regions that undergo phase separation in vitro and ii) are found within small, liquid-like puncta in cells that are proposed to form through phase separation. These dynamic puncta are proposed to drive co-localization of distal chromatin sites and organize the transcriptional machinery for coordinated expression of multiple genes. In the proposed studies, we will test the hypothesis that phase separation is a unifying mechanism that drives the formation of aberrant transcription centers by NUP98 FOs at chromatin sites targeted by their various C-terminal fused domains. These aberrant transcription centers, we propose, recruit components of the transcriptional machinery to activate expression of HOX and other genes and transform hematopoietic cells. 1