Jae Kyoung Kim

This paper introduces PINN-iPRC, a neural network framework that infers infinitesimal phase response curves from finite-pulse circadian recordings, achieving accurate results comparable to specialized experiments.

This paper derives and validates a fractional quasi-steady-state approximation for target-mediated drug disposition models with memory effects, showing improved model performance in adults but not infants.

This study determines the minimum sleep data duration needed for reliable circadian phase estimation and develops new methods to reduce this requirement.

This paper derives an exact formula for the competitive qTMDD model, improving computational speed and enabling accurate pharmacological parameter estimation in real-world clinical data.

The paper introduces a fully automated clustering pipeline for scRNA-seq that improves performance by 10-14% by integrating scLENS with existing tools.

This paper evaluates a novel pTMDD model for drug disposition using clinical data, showing improved accuracy and computational efficiency compared to existing models.

This paper introduces a noise-robust control mechanism for biomolecular systems that maintains output stability and reduces noise at the single-cell level.

The paper shows that combining three indirect repression mechanisms maintains ultrasensitivity and noise robustness in transcriptional switches under realistic biological conditions.

This study evaluates the accuracy of stochastic simulations of gene regulation under spatial homogeneity assumptions and finds that spatial heterogeneity must be considered for accurate results, though it increases computational cost.

This paper introduces a noise-resilient ultrasensitive transcriptional switch using indirect repression mechanisms and shows its effectiveness in noisy cellular environments.

The paper shows that noisy delay in genetic oscillators can improve temporal peak reliability and sharpen oscillation peaks without affecting the period.

The paper shows that proteins interact with the endoplasmic reticulum to accumulate near the nucleus without energy expenditure, following Chapmans law instead of Ficks law.

The paper develops Density-PINN to infer transduction-time distributions from single-cell gene expression data and finds that heterogeneity in response intensity is reduced when multiple signaling pathways regulate the response.

This paper introduces a general model-based causal inference method, GOBI, that overcomes limitations of model-free approaches by using ODEs to infer causality in dynamical systems.

The paper develops a mathematical model showing how spatially coordinated phosphorylation of PER protein ensures precise and robust circadian rhythms despite spatiotemporal noise.

This paper introduces a framework and tool to derive analytic steady states of complex biochemical reaction networks by breaking them into subnetworks and using a software package called COMPILES.

The paper develops a Bayesian MCMC method to infer gene regulatory network parameters from partial observations, demonstrating its ability to estimate delays and kinetic parameters in a two-step activation model.

The paper shows that combining multiple transcriptional repression mechanisms generates ultrasensitivity and oscillations, explaining their use in biological systems.

The paper shows that non-elementary propensities can accurately model stochastic dynamics of reversible bindings but fail in certain cases, and introduces a new framework with an open-source tool to address this.

The paper develops a computational package to analyze sleep-wake patterns and shows that aligning them with circadian rhythm reduces daytime sleepiness, even with less total sleep.

The paper develops a method to infer causality in oscillatory biological systems, outperforming existing methods and providing a computational tool for analysis.

The paper introduces a hierarchical Bayesian framework to infer variability in gene regulatory processes with delays using both simulated and experimental data, showing improved parameter estimates.

The paper develops a method to derive analytic stationary distributions for biochemical reaction networks by transforming them to be complex balanced, and introduces a computational tool for this purpose.

This study shows that mutating a PER2 phosphodegron in mice leads to longer circadian periods and altered molecular stability.

The paper introduces a Bayesian framework to infer reaction rates and delay distributions in biochemical networks using single-cell data, showing robust parameter estimation with limited measurements.

This study identifies MDM2 as a regulator of PER2 degradation and shows it influences circadian rhythm period length through a phosphorylation-independent mechanism.

This paper introduces a Bayesian method for estimating enzyme kinetic parameters with improved accuracy and efficiency compared to the traditional Michaelis-Menten approach.