Related Experiment Video
Updated: May 21, 2025
08:14
MicroRNA Based Liquid Biopsy: The Experience of the Plasma miRNA Signature Classifier MSC for Lung Cancer Screening
Published on: October 26, 2017
15.6K
Identification of hypertension subtypes using microRNA profiles and machine learning
Smarti Reel1, Parminder S Reel1, Josie Van Kralingen2
1Division of Population Health and Genomics, School of Medicine, University of Dundee, Dundee DD2 4BF, United Kingdom.
European Journal of Endocrinology
|March 19, 2025
Summary
This study shows that circulating microRNAs (miRNAs) can help distinguish endocrine hypertension (EHT) from primary hypertension (PHT). Machine learning identified specific miRNAs for accurate EHT subtype diagnosis.
Area of Science:
- Biomarker discovery
- Genomics
- Cardiovascular research
Background:
- Hypertension affects 1 in 3 adults, with primary hypertension (PHT) being most common.
- Endocrine hypertension (EHT), ~10% of cases, arises from conditions like primary aldosteronism (PA), Cushing's syndrome (CS), or pheochromocytoma/paraganglioma (PPGL).
- EHT is often misdiagnosed, delaying treatment and leading to ineffective management.
Purpose of the Study:
- To identify circulating microRNA (miRNA) biomarkers for distinguishing EHT and its subtypes from PHT.
- To evaluate the effectiveness of machine learning (ML) in classifying hypertension types using miRNA profiles.
Main Methods:
- Systematic analysis of circulating miRNA features.
- Application of 8 supervised machine learning methods for classification and prediction.
- Utilizing miRNA data to differentiate EHT subtypes (PA, CS, PPGL) from PHT.
Main Results:
- Machine learning models achieved high accuracy: AUC of 0.9 for classifying PPGL, CS, and EHT from PHT.
- Models achieved AUC of 0.8 for distinguishing PA from PHT.
- Key circulating miRNAs identified include hsa-miR-15a-5p and hsa-miR-32-5p.
Conclusions:
- Circulating miRNAs show significant potential as diagnostic biomarkers for EHT.
- Machine learning is a viable tool for identifying informative miRNAs for hypertension diagnosis.
