Computational Model Predicts Fed Batch Titer from Microplate Stress Profiles

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for predicting the fed batch production titer of producer cell clones are costly and time-consuming, as they rely on small-scale bioreactor systems to assess the performance of numerous clones, which can vary significantly in static culture versus fed-batch culture.

Innovation Solution

A rapid, high-throughput method involving microplate profiling of clonally-derived cells with chemical stressors to generate production titer response profiles, which are then used in a computational model to predict fed batch production titer, allowing for the ranking of clones for further development.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If small-scale bioreactor systems are used to measure clone performance, then prediction accuracy of fed-batch production titer is improved, but cost and time consumption increase

Engineering Contradiction:
Improveprediction accuracyVSAvoidtime consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates a computational model that copies the complex fed-batch culture behavior into a simplified static microplate assay model. By training the model on data from bioreactors and then using it to predict outcomes from simple microplate readings, it replicates the predictive accuracy of complex systems without requiring them for each screening event

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical bioreactor system with a computational modeling approach. Instead of physically running clones through expensive bioreactor systems, the invention uses a computer-based model that processes simple microplate data to predict fed-batch performance, substituting physical experimentation with computational prediction

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If small-scale bioreactor systems are used to assess clone performance, then prediction accuracy of fed-batch production titer is improved, but cost increases

Engineering Contradiction:
Improveprediction accuracyVSAvoidcost
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

The patent employs inexpensive, disposable 96-well microplates instead of expensive, reusable bioreactor systems. The microplates are simple, single-use items that can be rapidly processed, eliminating the high cost and complexity associated with bioreactor operation, maintenance, and sterilization while still providing sufficient data for accurate prediction when combined with the computational model

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

3Loss of time

If static microplate culture is used to assess clone performance, then time consumption is reduced, but prediction accuracy of fed-batch production titer deteriorates

Engineering Contradiction:
Improvetime consumptionVSAvoidprediction accuracy
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent introduces a computational model as an intermediary between the simple static microplate assay and the complex fed-batch culture outcome. The model acts as a translator that converts the simplified microplate data into accurate predictions of fed-batch performance, bridging the gap between the simplicity of the assay and the complexity of the target prediction

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3099807B1A method of predicting relative fed batch production titer of a panel of clonally-derived producer cells
Publication Date: 2019.02.13 VALITACELL
  • EP3099807B1 patent drawingFigure 1~2
  • EP3099807B1 patent drawingFigure 3~4
  • EP3099807B1 patent drawing

AI summary

"A rapid, high throughput, method of predicting relative fed batch performance of a panel of clonal cells derived from a single host cell line". A computer implemented method of predicting relative fed batch production titer of a panel of clonal producer cells derived from a single parental host cell population comprises the steps of assaying a level of growth of each clone in the presence and absence of at least three individual chemical cell stressors, comparing the level of growth of each clone in the presence and absence of the at least three chemical cell stressors to provide a normalised production titer response for each clone in each stressed microenvironment, and inputting a clone-specific production titer response profile comprising the normalised production titer response for each clone in each stressed microenvironments into a computational model. The computational model is generated from production titer response profiles obtained from a calibration set of clones with known fed batch production titer, and is configured to output the predicted relative fed batch production titer of the panel of clonal cells.