Enhancing generalizability of model discovery across parameter space with multi-experiment equation learning for biological systems
PLOS Computational Biology · 2026
DOI: 10.1371/journal.pcbi.1014161 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2026. Verfasst von Maria-Veronica Ciocanel und 5 weiteren Autor:innen. Beteiligte Einrichtung: Departments of Mathematics and Biology, Duke University, Durham, North Carolin, United States of America. Hauptthemen laut MEDLINE: Models, Biological, Machine Learning, Systems Biology. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: American Institute for Mathematics, NSF, Division of Intramural Research, National Institute of Allergy and Infectious Diseases.
Abstract (Originalsprache)
Agent-based modeling (ABM) is a powerful tool for understanding self-organizing biological systems, but it is computationally intensive and often not analytically tractable. Equation learning (EQL) methods can derive continuum models from ABM data, but they typically require extensive simulations for each parameter set, raising concerns about generalizability. In this work, we extend EQL to Multi-experiment equation learning (ME-EQL) by introducing two methods: (i) one-at-a-time ME-EQL (OAT ME-EQL), which learns individual models for each parameter set and connects them via interpolation, and (ii) embedded structure ME-EQL (ES ME-EQL), which builds a unified model library across parameters. We demonstrate these methods by learning continuum models from a noisy birth–death mean-field model and from an on-lattice agent-based model of birth, death, and migration with spatial structure, often used to investigate cell biology experiments. We show that both methods significantly reduce the relative error in recovering parameters from agent-based simulations, with OAT ME-EQL offering better generalizability across parameter space. Our findings highlight the potential of equation learning from multiple experiments to enhance the generalizability and interpretability of learned models for complex biological systems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Departments of Mathematics and Biology, Duke University, Durham, North Carolin, United States of America.
- Department of Mathematics and Statistics, The College of New Jersey, Ewing, New Jersey, United States of America.
- Department of Mathematics, Center for Research in Scientific Computation, North Carolina State University, Raleigh, North Carolina, United States of America.
- Department of Applied Mathematics, University of California Merced, Merced, California, United States of America.
- Department of Mathematics, Purdue University, West Lafayette, Indiana, United States of America.
ORCID-Profile: Maria-Veronica Ciocanel, John T Nardini, Kevin B Flores, Alexandria Volkening
Förderung
- American Institute for Mathematics
- NSF
- Division of Intramural Research, National Institute of Allergy and Infectious Diseases
Zitieren
APA
Ciocanel, M. V., Nardini, J. T., Flores, K. B., Rutter, E. M., Sindi, S. S., & Volkening, A. (2026). Enhancing generalizability of model discovery across parameter space with multi-experiment equation learning for biological systems. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1014161
Vancouver
Ciocanel MV, Nardini JT, Flores KB, Rutter EM, Sindi SS, Volkening A. Enhancing generalizability of model discovery across parameter space with multi-experiment equation learning for biological systems. PLOS Computational Biology. 2026. doi:10.1371/journal.pcbi.1014161
BibTeX
@article{Ciocanel2026,
author = {Ciocanel, Maria-Veronica and Nardini, John T and Flores, Kevin B and Rutter, Erica M and Sindi, Suzanne S and Volkening, Alexandria},
title = {Enhancing generalizability of model discovery across parameter space with multi-experiment equation learning for biological systems},
journal = {PLOS Computational Biology},
year = {2026},
doi = {10.1371/journal.pcbi.1014161}
}
EUVIMED