arXiv subjects

Keilung Choy

Publications and source records attributed to Keilung Choy.

6 recordsLinked to original sources

arXiv2026

Intracellular Measurement-Informed Multiscale Modeling for Scalable iPSC Manufacturing

Scalable manufacturing of human induced pluripotent stem cells (iPSCs) is essential for industrial-scale production of cell therapies and regenerative medicines. However, the 3D aggregate cultures used in manufacturing exhibit substantial spatial and metabolic heterogeneity compared with the relatively homogeneous monolayer systems used in laboratory studies, complicating mechanistic understanding and predictive metabolic modeling across culture scales. To address this challenge, we developed a modular multiscale mechanistic foundation model that links molecular, cellular, and macroscopic processes while accounting for spatial and metabolic heterogeneity. The framework integrates extracellular culture dynamics, intracellular metabolic fluxes, and cellular redox states by extending a previously established monolayer kinetic network and coupling it with a biological systems-of-systems (Bio-SoS) multiscale model for aggregate cultures, incorporating explicit redox interactions. Systematic monolayer and aggregate experiments (including multiple isotopic tracers, extracellular metabolite profiling, and two-photon optical redox imaging) were used to improve and validate the model. This integrated framework unifies heterogeneous datasets across culture configurations and enables mechanistic interpretation of metabolic and redox responses across heterogeneous culture scales, providing a quantitative foundation for scalable iPSC biomanufacturing.

q-bio.CB↗

arXiv2026

A Modular Mechanistic In Silico Model for In Vitro Transcription Process Yield and Product Quality Prediction

In vitro transcription (IVT) plays a critical role in the manufacture of mRNA vaccines and therapeutics. Optimizing mRNA yield and ensuring product quality, such as capping efficiency and integrity, are essential but mechanistically complex. This study presents a modular mechanistic model of the IVT process to advance scientific understanding and improve predictive capability. The IVT reaction network is decomposed into interconnected modules describing (1) initiation and capping, (2) elongation and truncation, (3) termination and read-through, (4) mRNA degradation, (5) magnesium pyrophosphate precipitation, and (6) enzymatic degradation of pyrophosphate. Guided by biochemical principles and experimental data, kinetic models were developed for each module, accounting for mass balances, molecular complexation, and enzyme activity, and were subsequently assembled to capture coupled IVT dynamics. Multivariate residual analysis and Shapley value-based sensitivity analysis, guided by domain knowledge, were applied to iteratively improve model fidelity. These machine learning-driven analytics enabled identification of key mechanisms, supported in silico experimentation, and facilitated root-cause analysis. Combined with Gaussian-process-based batch Bayesian optimization for efficient parameter estimation, this framework establishes a scalable hybrid (mechanistic + machine learning) modeling platform that integrates heterogeneous data, accelerates model calibration, and supports rational design and optimization of mRNA manufacturing processes.

q-bio.MN↗

arXiv2026

Variance Reduction Based Experience Replay for Policy Optimization

Effective reinforcement learning (RL) for complex stochastic systems requires leveraging historical data to improve sample efficiency and accelerate policy optimization. However, classical experience replay treats all past observations uniformly and fails to account for their varying contributions to learning. To address this limitation, we propose Variance Reduction Experience Replay (VRER), a principled framework that selectively reuses informative samples to reduce the variance of policy gradient estimates. VRER is algorithm-agnostic and can be integrated with existing policy optimization methods, yielding the sample-efficient off-policy algorithm, Policy Gradient with VRER (PG-VRER). To provide rigorous theoretical guarantees, we develop a novel analysis framework for experience replay that explicitly accounts for dependencies induced by Markovian dynamics and behavior-policy interactions. Using this framework, we establish finite-time convergence guarantees for PG-VRER and characterize a fundamental bias-variance trade-off: reusing older samples reduces gradient variance but may introduce greater estimation bias. Extensive experiments show that VRER consistently accelerates learning and outperforms state-of-the-art policy optimization algorithms

stat.ML↗

arXiv2025

A Symbolic and Statistical Learning Framework to Discover Bioprocessing Regulatory Mechanism: Cell Culture Example

Bioprocess mechanistic modeling is essential for advancing intelligent digital twin representation of biomanufacturing, yet challenges persist due to complex intracellular regulation, stochastic system behavior, and limited experimental data. This paper introduces a symbolic and statistical learning framework to identify key regulatory mechanisms and quantify model uncertainty. Bioprocess dynamics is formulated with stochastic differential equations characterizing intrinsic process variability, with a predefined set of candidate regulatory mechanisms constructed from biological knowledge. A Bayesian learning approach is developed, which is based on a joint learning of kinetic parameters and regulatory structure through a formulation of the mixture model. To enhance computational efficiency, a Metropolis-adjusted Langevin algorithm with adjoint sensitivity analysis is developed for posterior exploration. Compared to state-of-the-art Bayesian inference approaches, the proposed framework achieves improved sample efficiency and robust model selection. An empirical study demonstrates its ability to recover missing regulatory mechanisms and improve model fidelity under data-limited conditions.

stat.ML↗

arXiv2025

Digital Twin Calibration with Model-Based Reinforcement Learning

This paper presents a novel methodological framework, called the Actor-Simulator, that incorporates the calibration of digital twins into model-based reinforcement learning for more effective control of stochastic systems with complex nonlinear dynamics. Traditional model-based control often relies on restrictive structural assumptions (such as linear state transitions) and fails to account for parameter uncertainty in the model. These issues become particularly critical in industries such as biopharmaceutical manufacturing, where process dynamics are complex and not fully known, and only a limited amount of data is available. Our approach jointly calibrates the digital twin and searches for an optimal control policy, thus accounting for and reducing model error. We balance exploration and exploitation by using policy performance as a guide for data collection. This dual-component approach provably converges to the optimal policy, and outperforms existing methods in extensive numerical experiments based on the biopharmaceutical manufacturing domain.

cs.LG↗

arXiv2024

Adjoint Sensitivity Analysis on Multi-Scale Bioprocess Stochastic Reaction Network

Motivated by the pressing challenges in the digital twin development for biomanufacturing systems, we introduce an adjoint sensitivity analysis (SA) approach to expedite the learning of mechanistic model parameters. In this paper, we consider enzymatic stochastic reaction networks representing a multi-scale bioprocess mechanistic model that allows us to integrate disparate data from diverse production processes and leverage the information from existing macro-kinetic and genome-scale models. To support forward prediction and backward reasoning, we develop a convergent adjoint SA algorithm studying how the perturbations of model parameters and inputs (e.g., initial state) propagate through enzymatic reaction networks and impact on output trajectory predictions. This SA can provide a sample efficient and interpretable way to assess the sensitivities between inputs and outputs accounting for their causal dependencies. Our empirical study underscores the resilience of these sensitivities and illuminates a deeper comprehension of the regulatory mechanisms behind bioprocess through sensitivities.

q-bio.MN↗