EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

Gaussian Processes for Predictive QSAR Modeling of Chromatographic Processes

Harini Narayanan iD, Douglas Nolan, Lijuan Li, Edward King, Andrew Fiordalis, David Benjamin Nickel, George Parks iD, J Christopher Love iD

Biotechnology and Bioengineering · 2026

DOI: 10.1002/bit.70168 ↗

At a glance

Journal article in Biotechnology and Bioengineering, 2026. Written by Harini Narayanan and 7 co-authors. Institution: Koch Institute for Integrative Cancer Research at MIT, Cambridge, Massachusetts, USA. Major MEDLINE topics: Quantitative Structure-Activity Relationship, Chromatography, Machine Learning. 27 references. The full text is freely available (open access, CC BY 4.0). Funded by: Takeda Development Center Americas, Inc., Massachusetts Institute of Technology.

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Chromatography is a key unit operation in the biopharmaceutical manufacturing process used for protein purification and polishing. Design and optimization of these processes are resource‐intensive resulting from the complex combinatorial design space. This constraint combined with the wide diversity in therapeutic formats and increased pressure for timely delivery to the market necessitates an efficient, fast, robust and generalized framework for process design and optimization. Here we present Gaussian processes as a potent machine learning methodology for predictive modeling in the context of QSAR modeling used for resin and solvent condition selection. We highlight the on‐par predictive power of Gaussian Processes with other reported machine learning algorithms. Furthermore, we demonstrate additional properties of Gaussian processes such as its ability to provide confidence estimates for its prediction that makes it suitable for model‐assisted optimization. Finally, we demonstrate the possibility to derive feature importances from Gaussian processes, making these models as interpretable as ensembled tree‐based methods such as random forests.

Subjects

MeSH terms (MEDLINE)

Quantitative Structure-Activity RelationshipChromatographyMachine LearningNormal DistributionPredictive Learning ModelsPrediction Algorithms

Author keywords

Bayesian approachesbiopharmaceutical manufacturingbiotechnologyhydrophobic interaction chromatographypolishing

Authors and institutions

  • Koch Institute for Integrative Cancer Research at MIT, Cambridge, Massachusetts, USA.
  • Takeda Development Center Americas, Lexington, Massachusetts, USA.
  • Baxalta Innovations GmbH, part of Takeda, Vienna, Austria.
  • Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.

ORCID profiles: Harini Narayanan, George Parks, J Christopher Love

Funding

  • Takeda Development Center Americas, Inc.
  • Massachusetts Institute of Technology

Cite

APA

Narayanan, H., Nolan, D., Li, L., King, E., Fiordalis, A., Nickel, D. B., Parks, G., & Love, J. C. (2026). Gaussian Processes for Predictive QSAR Modeling of Chromatographic Processes. Biotechnology and Bioengineering. https://doi.org/10.1002/bit.70168

Vancouver

Narayanan H, Nolan D, Li L, King E, Fiordalis A, Nickel DB, et al. Gaussian Processes for Predictive QSAR Modeling of Chromatographic Processes. Biotechnology and Bioengineering. 2026. doi:10.1002/bit.70168

BibTeX
@article{Narayanan2026,
  author = {Narayanan, Harini and Nolan, Douglas and Li, Lijuan and King, Edward and Fiordalis, Andrew and Nickel, David Benjamin and Parks, George and Love, J Christopher},
  title = {Gaussian Processes for Predictive QSAR Modeling of Chromatographic Processes},
  journal = {Biotechnology and Bioengineering},
  year = {2026},
  doi = {10.1002/bit.70168}
}
Start for free →