Hybrid Cell Metabolism Observer for Bioreactor State Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biomanufacturing processes lack direct measurements of cell metabolism, leading to inefficient and ad hoc optimization of bioprocesses, difficulty in diagnosing abnormal metabolic operations, and limited understanding of metabolic states, resulting in suboptimal productivity and high byproduct formation.
Innovation Solution
A computer-implemented method using a hybrid model that includes a kinetic growth model and a metabolic condition model to predict and classify metabolic states, enabling real-time observability and optimization of bioprocess conditions, with features like machine learning and statistical modeling to enhance prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If indirect measurements of average cell metabolic behavior are used, then process monitoring is simplified, but measurement precision of actual metabolic activity is insufficient
Solution Approach 1:
The patent introduces an observer system as an intermediary that uses measurable outputs (metabolite consumption/production rates) to infer unmeasured internal states (metabolic activity). This mediator translates indirect measurements into meaningful metabolic information without requiring direct measurement of cellular processes.
Solution Approach 2:
The patent replaces direct mechanical/biological measurement of metabolic activity with a computational modeling approach. Instead of physically measuring metabolic fluxes, the system uses mathematical models and optimization algorithms to calculate metabolic states from observable process data.
2Measurement precision
If flux balance analysis with genetic knowledge is used, then model accuracy improves, but device complexity and data requirements increase significantly
Solution Approach 1:
The patent changes the parameters used in metabolic modeling from requiring detailed genetic information and gene-expression data to using readily available process data such as metabolite consumption and production rates. This parameter transformation simplifies the model while maintaining predictive accuracy for bioprocess optimization.
Solution Approach 2:
The patent applies partial action by focusing the model only on the essential metabolic fluxes relevant to bioprocess performance, rather than attempting to model the complete cellular metabolic network. This selective approach reduces model complexity while capturing the critical metabolic behavior needed for process optimization.
3Measurement precision
If statistical regression models are used, then predictive capability is achieved, but understanding of metabolic function and process variables is limited
Solution Approach 1:
The patent implements feedback by using the observer system to continuously monitor and provide information about the actual metabolic state of the bioprocess. This feedback loop enables real-time understanding of metabolic function, allowing operators to adjust process conditions based on actual metabolic performance rather than relying solely on predictive correlations.
Solution Approach 2:
The observer system acts as an intermediary that bridges the gap between statistical prediction and metabolic understanding. It translates process data into meaningful metabolic information, providing both predictive capability and insight into the underlying metabolic mechanisms driving bioprocess performance.
4Reliability
If traditional fed-batch processes operate at safe conditions with abundant nutrients, then process reliability is maintained, but productivity and byproduct formation are suboptimal
Solution Approach 1:
The patent applies dynamics by transitioning from static, conservative feeding strategies to dynamic feeding control based on real-time metabolic state observation. The system continuously adjusts nutrient supply according to the actual metabolic needs of the cells, enabling the process to operate optimally across different phases while maintaining stability through active control.
Solution Approach 2:
The observer system provides feedback on the actual metabolic state, enabling closed-loop control of feeding strategies. This feedback allows the system to push operating conditions closer to optimal productivity regions while maintaining process reliability through real-time monitoring and adjustment, rather than relying on conservative open-loop control.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Techniques for predicting an amount of at least one biomaterial produced or consumed by a biological system in a bioreactor are provided. Process conditions and metabolite concentrations are measured for the biological system as a function of time. Metabolic rates for the biological system, including specific consumption rates of metabolites and specific production rates of metabolites are determined. The process conditions and the metabolic rates are provided to a hybrid system model configured to predict production of the biomaterial. The hybrid system model includes a kinetic growth model configured to estimate cell growth as a function of time and a metabolic condition model based on metabolite specific consumption or secretion rates and select process conditions, wherein the metabolic condition model is configured to classify the biological system into a metabolic state. An amount of the biomaterial based on the hybrid system model is predicted.