KAIST Succeeds in Restoring Altered Cellular Stimulus-Response Patterns to Normal

A universal technology that identifies gene control targets within altered gene networks in cells and restores them to a normal state has been developed in South Korea. This technology can be applied to novel cancer therapies such as cancer reversibility, new drug development, precision medicine, and cell therapy reprogramming.

KAIST announced on August 28 that the research team led by Professor Kwanghyun Cho from the Department of Bio and Brain Engineering has developed a technology to identify "gene control targets" that restore the stimulus-response patterns of altered cells to normal using an algebraic approach.

(From left) Master Insoo Jung, PhD candidates Corbin Harper, Sunghoon Jang, Hyunsoo Yeo, Professor Kwanghyun Cho. Provided by KAIST

The algebraic approach expresses gene networks as mathematical equations and uses algebraic calculations to identify control targets.

The research team represented the complex relationships where genes regulate each other within a cell as a single "Boolean network" (logical circuit diagram).

They also visualized how cells respond to external stimuli as a "phenotype landscape" and applied a mathematical technique called "semi-tensor product" to analyze and determine which genes to regulate, how the overall cellular response would change, and how to calculate this rapidly and accurately. The semi-tensor product refers to calculating all possible gene combinations and control effects as a single algebraic formula.

However, in this process, the number of key genes that determine the actual cellular response exceeds several thousand, making the calculations extremely complex. To address this, the research team also applied a "numerical approximation (Taylor approximation)" technique to simplify the calculations. Numerical approximation allows complex problems to be solved with simpler formulas while yielding similar results.

Through this, the team succeeded in calculating which "stable state (attractor)" the cell reaches and predicting how the cell would transition to a new state when specific genes are controlled. This was made possible by identifying core gene control targets that can revert abnormal cellular responses to a state most similar to normal.

For example, the research team applied this technology to the gene network of bladder cancer cells and identified gene control targets that restore the altered responses to normal. They also found gene control targets that could restore normal stimulus-response patterns in immune cells with massively distorted gene networks during differentiation. This is expected to provide a foundation for rapidly and systematically solving problems that previously could only be explored approximately through lengthy computer simulations.

Professor Kwanghyun Cho stated, "This research is a core foundational technology for developing the Digital Cell Twin model, which analyzes and controls the phenotype landscape of gene networks that determine cell fate. This technology can be widely applied across the life sciences and medicine, including new cancer therapies through cancer reversibility and new drug development reprogramming."

The Digital Cell Twin model is a technology that digitally models the complex reaction processes occurring inside cells, replacing experiments by simulating cellular responses in a virtual environment.

Meanwhile, this research involved Master Insoo Jung, PhD candidates Corbin Harper, Sunghoon Jang, and Hyunsoo Yeo from KAIST. The research results (paper) were published online in the international journal Science Advances, published by the American Association for the Advancement of Science (AAAS), on August 22.

This content was produced with the assistance of AI translation services.