Our first paper came out in Molecular Cell today! This is Shasha’s work together with the Tjian/Darzacq Lab at UC Berkeley. We report that endogenous gene transcription requires narrow optimum of transcription factor (TF) low-complexity domain (LCD) multivalent interactions, and phase separation can be counterproductive (A). We also show that mislocalized TF LCD multivalent interactions can outcompete stochiometric TF-DNA binding and sequester a TF away from its target genes to distrupt its functions, indicating a new therapeutic strategy for targeting disease-causing TFs, e.g., EWS::FLI1 (B). Furthermore, we provide proof-of-concept demonstration that single-molecule measurements of protein diffusion rates can serve as an effective means for detecting phase separation in vivo. We detected a slower TF diffusion in the phase-separated nucleolus than in the nucleoplasm (C). Check out the details in "Tuning levels of low-complexity domain interactions to modulate endogenous oncogenic transcription."