Noise-Guided Design of Synthetic Protein Waves in Living Cells

Protein circuits organize cell biology, but synthetic dynamics are challenging to engineer due to stochastic genetic and biochemical variation. Genetically encoded oscillators (GEOs) built from bacterial MinDE-family ATPases and activators generate synthetic protein waves that act as novel frequency-domain imaging barcodes in eukaryotic cells, providing a platform for understanding, engineering, and applying synthetic protein dynamics. Using budding yeast, we disentangle how expression levels and expression noise govern the GEO waveform and encodability. While the GEO amplitude is sensitive to extrinsic noise, the GEO frequency is stably encoded by the activator:ATPase ratio. By integrating GEO components into the yeast modular cloning toolkit, we developed different noise-guided expression strategies that act like filters on the GEO waveform. We paired these filters with hundreds of biochemically distinct GEO variants to engineer clonal populations that oscillate at distinct frequencies and to design waveform libraries with customizable spectral features and tunable waveform variation. Our work establishes a robust platform for precision genetic encoding of synthetic GEO oscillations and highlights the utility of noise-guided strategies for dynamic protein circuit design.