Machine Learning‐Driven Nanopore Sensing for Quantitative, Label‐Free miRNA Detection
Small Methods · 2026
DOI: 10.1002/smtd.202502335 ↗Kurz gefasst
Fachartikel in der Zeitschrift Small Methods, 2026. Verfasst von Caroline Koch und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry, Molecular Science Research Hub, Imperial College London, London, UK. Hauptthemen laut MEDLINE: MicroRNAs, Nanopores, Machine Learning, Biosensing Techniques. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Analytical Chemistry Trust Fund, H2020 European Research Council, Cancer Research Institute.
Abstract (Originalsprache)
Nanopore sensors offer exceptional sensitivity for detecting single molecules, making them ideal for early disease diagnostics. In this study, we present a multiplexed nanopore‐based assay that combines DNA‐barcoded probes with advanced computational analysis to detect microRNAs (miRNAs) with high specificity and accuracy. Each probe selectively binds its target biomarker and induces a characteristic delay in the ionic current signal upon translocation through the nanopore. We evaluated three analytical strategies for classifying delayed versus non‐delayed events: (1) moving standard deviation (MSD), (2) spectral entropy (SE), and (3) a convolutional neural network (CNN). While MSD and SE rely on manually defined thresholds and exhibit limited sensitivity, the CNN model, trained on image representations of raw current traces, achieved near‐perfect classification performance across all metrics (accuracy = 0.99, precision = 0.99, recall = 0.99). Grad‐CAM visualization confirmed that the CNN model focused on relevant signal regions, enhancing interpretability and generalizability. All methods produced sigmoidal concentration‐response curves consistent with expected binding kinetics, and nanopore‐derived delay metrics closely matched RT‐qPCR validation data. All three methods were capable of distinguishing between signal classes; however, the CNN model demonstrated superior sensitivity and robustness. This work highlights the importance of data interpretation in nanopore sensing and presents a comparative framework for binary event classification. The findings pave the way for the development of machine learning‐driven nanopore diagnostics capable of detecting diverse biomarker types at the single‐molecule level.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Chemistry, Molecular Science Research Hub, Imperial College London, London, UK.
- Department of Life Science, Sir Alexander Fleming Building, Imperial College London, London, UK.
- Oxford Nanopore Technologies, Oxford, UK.
- National Heart and Lung Institute, ICTEM, Imperial College London, London, UK.
ORCID-Profile: Seshagiri Sakthimani, Julia Gorelik, Nadia Guerra, Joshua B Edel, Aleksandar P Ivanov
Förderung
- Analytical Chemistry Trust Fund
- H2020 European Research Council
- Cancer Research Institute
Zitieren
APA
Koch, C., Sakthimani, S., Noakes, V. M., Cretu, M., Newman, D., Gutierrez, R., Bruce, M., Gorelik, J., Guerra, N., Edel, J. B., & Ivanov, A. P. (2026). Machine Learning‐Driven Nanopore Sensing for Quantitative, Label‐Free miRNA Detection. Small Methods. https://doi.org/10.1002/smtd.202502335
Vancouver
Koch C, Sakthimani S, Noakes VM, Cretu M, Newman D, Gutierrez R, et al. Machine Learning‐Driven Nanopore Sensing for Quantitative, Label‐Free miRNA Detection. Small Methods. 2026. doi:10.1002/smtd.202502335
BibTeX
@article{Koch2026,
author = {Koch, Caroline and Sakthimani, Seshagiri and Noakes, Victoria Maria and Cretu, Miruna and Newman, David and Gutierrez, Richard and Bruce, Mark and Gorelik, Julia and Guerra, Nadia and Edel, Joshua B and Ivanov, Aleksandar P},
title = {Machine Learning‐Driven Nanopore Sensing for Quantitative, Label‐Free miRNA Detection},
journal = {Small Methods},
year = {2026},
doi = {10.1002/smtd.202502335}
}
EUVIMED