EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Leveraging synthetic and genetic data to improve epidemic forecasting

Dave Osthus iD, Alexander C Murph iD, Emma E Goldberg, Lauren J Beesley, William M Fischer, Nidhi Parikh, Lauren A Castro

PLOS Computational Biology · 27.08.2026 · 22 (8) · e1014630

DOI: 10.1371/journal.pcbi.1014630 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 27.08.2026. Verfasst von Dave Osthus und 6 weiteren Autor:innen. Beteiligte Einrichtung: Statistics Group, Los Alamos National Laboratory, Los Alamos, New Mexico, United States of America. Hauptthemen laut MEDLINE: Epidemics. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Laboratory Directed Research and Development program, Los Alamos National Laboratory.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Forecasting infectious disease outbreaks is hard. Forecasting emerging infectious diseases with limited historical data is even harder. In this paper, we investigate ways to improve emerging infectious disease forecasting when little pathogen-specific training data are available. Specifically, we explore two sources of information that may be available near the start of an emerging disease outbreak: synthetic data and genetic information. For this investigation, we conducted an experiment where we trained deep learning models on different combinations of real and synthetic data, both with and without genetic information, to explore how these models compare when forecasting COVID-19 cases for US states. All models are developed with an eye towards forecasting the next pandemic. We find that models trained with synthetic data have better forecast accuracy than models trained on real data alone, and models that use genetic variants have better forecast accuracy compared to those that do not. All models outperformed a baseline persistence model, a benchmark that proved challenging for many real-time COVID-19 case forecasting models, and multiple models outperformed the COVIDHub-4_week_ensemble. This paper demonstrates the value of these underutilized sources of information and provides a blueprint for forecasting future pandemics.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

EpidemicsCOVID-19ForecastingHumansSARS-CoV-2PandemicsDeep LearningComputational BiologyPredictive Learning ModelsPrediction AlgorithmsUnited StatesDisease OutbreaksCommunicable Diseases, Emerging

Autor:innen und Institutionen

  • Statistics Group, Los Alamos National Laboratory, Los Alamos, New Mexico, United States of America.
  • Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, New Mexico, United States of America.
  • Computational Intelligence and Modeling Group, Los Alamos National Laboratory, Los Alamos, New Mexico, United States of America.

ORCID-Profile: Dave Osthus, Alexander C Murph

Förderung

  • Laboratory Directed Research and Development program
  • Los Alamos National Laboratory

Zitieren

APA

Osthus, D., Murph, A. C., Goldberg, E. E., Beesley, L. J., Fischer, W. M., Parikh, N., & Castro, L. A. (2026). Leveraging synthetic and genetic data to improve epidemic forecasting. PLOS Computational Biology, 22(8), e1014630. https://doi.org/10.1371/journal.pcbi.1014630

Vancouver

Osthus D, Murph AC, Goldberg EE, Beesley LJ, Fischer WM, Parikh N, et al. Leveraging synthetic and genetic data to improve epidemic forecasting. PLOS Computational Biology. 2026;22(8):e1014630. doi:10.1371/journal.pcbi.1014630

BibTeX
@article{Osthus2026,
  author = {Osthus, Dave and Murph, Alexander C and Goldberg, Emma E and Beesley, Lauren J and Fischer, William M and Parikh, Nidhi and Castro, Lauren A},
  title = {Leveraging synthetic and genetic data to improve epidemic forecasting},
  journal = {PLOS Computational Biology},
  year = {2026},
  volume = {22},
  number = {8},
  pages = {e1014630},
  doi = {10.1371/journal.pcbi.1014630}
}
Kostenlos starten →