EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

Interpretable‐AI‐Based Model Structural Transfer Learning to Accelerate Bioprocess Model Construction

Alexander W Rogers iD, Fernando Vega-Ramon, Amanda Lane, Philip Martin, Dongda Zhang iD

Biotechnology and Bioengineering · 2025-10 · 122 (10) · 2819-2831

DOI: 10.1002/bit.70026 ↗

At a glance

Journal article in Biotechnology and Bioengineering, published online on 2025-07-18. Written by Alexander W Rogers and 4 co-authors. Institution: Department of Chemical Engineering, The University of Manchester, Manchester, UK. Major MEDLINE topics: Machine Learning, Models, Biological. Cited 15 times so far (Crossref); 48 references. The full text is freely available (open access, CC BY 4.0). Funded by: This study was supported by the Engineering and Physical Sciences Research Council (EPSRC) and Unilever., Engineering and Physical Sciences Research Council.

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Determining accurate kinetic models for new biochemical systems is time‐intensive, requiring experimental data collection, model construction, validation, and discrimination. Traditional black‐box machine learning‐based transfer learning methods leverage prior knowledge but lack interpretability and physical insights. To address this, we propose a novel model structural transfer learning approach that combines symbolic regression with artificial neural network feature attribution. The method enables automatic structural modification of an inaccurate or low‐fidelity mechanistic model developed for one system when being applied to another system. Through a comprehensive in silico case study, our framework successfully adapted a kinetic model from one biochemical system to a different but related one, improving predictive accuracy. Moreover, the framework can significantly accelerate model identification when being integrated with model‐based design of experiments. By comparing the old and new model structures, physical insight can be obtained, altogether highlighting the framework's potential for advancing automated knowledge discovery and facilitating high‐fidelity predictive digital twin design for novel biochemical processes.

Subjects

MeSH terms (MEDLINE)

Machine LearningModels, BiologicalNeural Networks, ComputerComputer Simulation

Author keywords

bioprocess kineticsdigital twininterpretable machine learningknowledge discoverymodel‐based design of experiments

Authors and institutions

  • Department of Chemical Engineering, The University of Manchester, Manchester, UK.
  • Unilever R&D Port Sunlight, Wirral, Liverpool, UK.

ORCID profiles: Alexander W Rogers, Dongda Zhang

Funding

  • This study was supported by the Engineering and Physical Sciences Research Council (EPSRC) and Unilever.
  • Engineering and Physical Sciences Research Council

Cite

APA

Rogers, A. W., Vega-Ramon, F., Lane, A., Martin, P., & Zhang, D. (2025). Interpretable‐AI‐Based Model Structural Transfer Learning to Accelerate Bioprocess Model Construction. Biotechnology and Bioengineering, 122(10), 2819-2831. https://doi.org/10.1002/bit.70026

Vancouver

Rogers AW, Vega-Ramon F, Lane A, Martin P, Zhang D. Interpretable‐AI‐Based Model Structural Transfer Learning to Accelerate Bioprocess Model Construction. Biotechnology and Bioengineering. 2025;122(10):2819-2831. doi:10.1002/bit.70026

BibTeX
@article{Rogers2025,
  author = {Rogers, Alexander W and Vega-Ramon, Fernando and Lane, Amanda and Martin, Philip and Zhang, Dongda},
  title = {Interpretable‐AI‐Based Model Structural Transfer Learning to Accelerate Bioprocess Model Construction},
  journal = {Biotechnology and Bioengineering},
  year = {2025},
  volume = {122},
  number = {10},
  pages = {2819--2831},
  doi = {10.1002/bit.70026}
}
Start for free →