Cancer relapse after chemotherapy and bacterial persistence after antibiotics often trace back to a small fraction of cells that behave differently from the rest. These differences do not always arise from genetic changes. Random fluctuations in molecular activity, often described as biological noise, can push genetically identical cells into rare but consequential states.
A team led by Kim Jae Kyoung at KAIST and the IBS Biomedical Mathematics Group, Kim Jinsu at POSTECH, and Cho Byung-Kwan at KAIST now reports a mathematical framework designed to control such fluctuations at the single-cell level. Their work introduces a strategy that regulates variability itself, rather than focusing only on population averages.
Cells maintain homeostasis through feedback, but classical control strategies in synthetic biology stabilize mean protein levels while allowing wide swings in individual cells. The authors liken this to an everyday frustration. “Standard control methods are like adjusting a shower,” they explained. “You might get the water to average 40°C, but if that average is achieved by alternating between freezing cold and boiling hot water, you can’t take a shower.”
The new framework, termed the Noise Controller, is built on mathematical modeling of gene regulatory circuits. Instead of sensing only protein abundance, the controller responds to the magnitude of fluctuations, mathematically defined as the second moment of protein levels. The design relies on protein dimerization paired with targeted degradation, allowing cells to damp their own noise through feedback.
Simulations tested the approach in the DNA repair network of Escherichia coli. In standard models, about 20 percent of cells failed to activate repair pathways due to stochastic variation. With the Noise Controller applied, the failure rate dropped to 7 percent, while average protein levels remained stable. The results show that stochastic fluctuations can be constrained to near the physical limit described by a Fano factor of 1.
Publication Details
Lim, D., Moon, S., Song, Y.M. et al. Toward single-cell control: noise-robust perfect adaptation in biomolecular systems. Nat Commun (2025). https://doi.org/10.1038/s41467-025-67736-y
A team led by Kim Jae Kyoung at KAIST and the IBS Biomedical Mathematics Group, Kim Jinsu at POSTECH, and Cho Byung-Kwan at KAIST now reports a mathematical framework designed to control such fluctuations at the single-cell level. Their work introduces a strategy that regulates variability itself, rather than focusing only on population averages.
Cells maintain homeostasis through feedback, but classical control strategies in synthetic biology stabilize mean protein levels while allowing wide swings in individual cells. The authors liken this to an everyday frustration. “Standard control methods are like adjusting a shower,” they explained. “You might get the water to average 40°C, but if that average is achieved by alternating between freezing cold and boiling hot water, you can’t take a shower.”
The new framework, termed the Noise Controller, is built on mathematical modeling of gene regulatory circuits. Instead of sensing only protein abundance, the controller responds to the magnitude of fluctuations, mathematically defined as the second moment of protein levels. The design relies on protein dimerization paired with targeted degradation, allowing cells to damp their own noise through feedback.
Simulations tested the approach in the DNA repair network of Escherichia coli. In standard models, about 20 percent of cells failed to activate repair pathways due to stochastic variation. With the Noise Controller applied, the failure rate dropped to 7 percent, while average protein levels remained stable. The results show that stochastic fluctuations can be constrained to near the physical limit described by a Fano factor of 1.
Publication Details
Lim, D., Moon, S., Song, Y.M. et al. Toward single-cell control: noise-robust perfect adaptation in biomolecular systems. Nat Commun (2025). https://doi.org/10.1038/s41467-025-67736-y
