EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Exploring the application of deep learning methods for polygenic risk score estimation

Steven Squires iD, Michael N Weedon, Richard A Oram iD

Biomedical Physics & Engineering Express · 2025

DOI: 10.1088/2057-1976/adbb71 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomedical Physics & Engineering Express, 2025. Verfasst von Steven Squires und 2 weiteren Autor:innen. Beteiligte Einrichtung: Clinical and Biomedical Sciences, Faculty of Health and Life Sciences, University of Exeter, Exeter, United Kingdom. Hauptthemen laut MEDLINE: Deep Learning, Multifactorial Inheritance, Genetic Predisposition to Disease. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Randox Ltd, National Institute for Health and Care Research Exeter Biomedical Research Centre.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Polygenic risk scores (PRS) summarise genetic information into a single number with clinical and research uses. Deep learning (DL) has revolutionised multiple fields, however, the impact of DL on PRSs has been less significant. We explore how DL can improve the generation of PRSs.

Methods

We train DL models on known PRSs using UK Biobank data. We explore whether the models can recreate human programmed PRSs, including using a single model to generate multiple PRSs, and DL difficulties in PRS generation. We investigate how DL can compensate for missing data and constraints on performance.

Results

We demonstrate almost perfect generation of multiple PRSs with little loss of performance with reduced quantity of training data. For an example set of missing SNPs the DL model produces predictions that enable separation of cases from population samples with an area under the receiver operating characteristic curve of 0.847 (95% CI: 0.828–0.864) compared to 0.798 (95% CI: 0.779–0.818) for the PRS.

Conclusions

DL can accurately generate PRSs, including with one model for multiple PRSs. The models are transferable and have high longevity. With certain missing SNPs the DL models can improve on PRS generation; further improvements would likely require additional input data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningMultifactorial InheritanceGenetic Predisposition to DiseaseHumansPolymorphism, Single NucleotideROC CurveRisk AssessmentUnited KingdomGenome-Wide Association StudyRisk FactorsModels, GeneticAlgorithmsGenetic Risk Score

Stichwörter der Autor:innen

deep learninggeneticsmachine learningpolygenic risk scoresprecision medicine

Autor:innen und Institutionen

  • Clinical and Biomedical Sciences, Faculty of Health and Life Sciences, University of Exeter, Exeter, United Kingdom.
  • The Academic Renal Unit, Royal Devon University Healthcare Foundation Trust, United Kingdom.

ORCID-Profile: Steven Squires, Richard A Oram

Förderung

  • Randox Ltd
  • National Institute for Health and Care Research Exeter Biomedical Research Centre

Zitieren

APA

Squires, S., Weedon, M. N., & Oram, R. A. (2025). Exploring the application of deep learning methods for polygenic risk score estimation. Biomedical Physics & Engineering Express. https://doi.org/10.1088/2057-1976/adbb71

Vancouver

Squires S, Weedon MN, Oram RA. Exploring the application of deep learning methods for polygenic risk score estimation. Biomedical Physics & Engineering Express. 2025. doi:10.1088/2057-1976/adbb71

BibTeX
@article{Squires2025,
  author = {Squires, Steven and Weedon, Michael N and Oram, Richard A},
  title = {Exploring the application of deep learning methods for polygenic risk score estimation},
  journal = {Biomedical Physics & Engineering Express},
  year = {2025},
  doi = {10.1088/2057-1976/adbb71}
}
Kostenlos starten →