Identifying potential circulating miRNA biomarkers for the diagnosis and prediction of ovarian cancer using machine-learning approach: application of Boruta

Aug 5, 2026

Ovarian cancer poses a major clinical challenge due to limited characteristic symptoms and insufficient noninvasive biomarkers, leading to many diagnoses at advanced stages. This study uses Boruta, a random-forest feature-selection method, to identify miRNAs linked to ovarian cancer from GEO dataset GSE106817, with external validation in GSE113486 and GSE113740. Multiple machine-learning classifiers evaluate diagnostic performance, and ten miRNAs distinguish cases from controls with high reported AUCs, supporting miRNAs as screening candidates.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Methods

Results

Keywords

Background and Epidemiology

TYPE Original Research PUBLISHED 09 August 2023 DOI 10.3389/fdgth.2023.1187578 EDITED BY Ayan Banerjee, Arizona State University, United States REVIEWED BY Saurav Mallik, Harvard University, United States Imane Lamrani, Nikola Corporation, United States Payal Kamboj, Arizona State University, United States, in collaboration with reviewer [IL] *CORRESPONDENCE Neda Gilani [neda.gilani@gmail.com](neda.gilani@gmail.com)[ ](neda.gilani@gmail.com)Reza Arabi Belaghi [rezaarabi11@gmail.com](rezaarabi11@gmail.com)[ ](rezaarabi11@gmail.com)RECEIVED 16 March 2023 ACCEPTED 20 July 2023 PUBLISHED 09 August 2023 CITATION Hamidi F, Gilani N, Arabi Belaghi R, Yaghoobi H, Babaei E, Sarbakhsh P and Malakouti J (2023) Identifying potential circulating miRNA biomarkers for the diagnosis and prediction of ovarian cancer using machine-learning approach: application of Boruta. Front. Digit. Health 5:1187578 . doi: 10.3389/fdgth.2023.1187578 COPYRIGHT © 2023 Hamidi, Gilani, Arabi Belaghi, Yaghoobi, Babaei, Sarbakhsh and Malakouti. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Identifying potential circulating miRNA biomarkers for the diagnosis and prediction of ovarian cancer using machine-learning approach: application of Boruta Farzaneh Hamidi1, Neda Gilani1,2*, Reza Arabi Belaghi3,4,5*, Hanif Yaghoobi6, Esmaeil Babaei6,7, Parvin Sarbakhsh1 and Jamileh Malakouti8 1Department of Statistics and Epidemiology, Faculty of Health, Tabriz University of Medical Sciences, Tabriz, Iran, 2Road Trafﬁc Injury Research Center, Tabriz University of Medical Sciences, Tabriz, Iran, 3Department of Mathematics, Applied Mathematics and Statistics, Uppsala University, Uppsala, Sweden, 4Department of Statistics, Faculty of Mathematical Science, University of Tabriz, Tabriz, Iran, 5Department of Energy and Technology, Swedish Agricultural University, Uppsala, Sweden, 6Department of Biological Sciences, School of Natural Sciences, University of Tabriz, Tabriz, Iran, 7Interfaculty Institute for Bioinformatics and Medical Informatics (IBMI), University of Tübingen, Tübingen, Germany, 8Department of Midwifery, Faculty of Nursing and Midwifery, Tabriz University of Medical Science, Tabriz, Iran Introduction: In gynecologic oncology, ovarian cancer is a great clinical challenge. Because of the lack of typical symptoms and effective biomarkers for noninvasive screening, most patients develop advanced-stage ovarian cancer by the time of diagnosis. MicroRNAs (miRNAs) are a type of non-coding RNA molecule that has been linked to human cancers. Specifying diagnostic biomarkers to determine non-cancer and cancer samples is difﬁcult. Methods: By using Boruta, a novel random forest-based feature selection in the machine-learning techniques, we aimed to identify biomarkers associated with ovarian cancer using cancerous and non-cancer samples from the Gene Expression Omnibus (GEO) database: GSE106817 . In this study, we used two independent GEO data sets as external validation, including GSE113486 and GSE113740 . We utilized ﬁve state-of-the-art machine-learning algorithms for classiﬁcation: logistic regression, random forest, decision trees, artiﬁcial neural networks, and XGBoost. Results: Four models discovered in GSE113486 had an AUC of 100%, three in GSE113740 with AUC of over 94%, and four in GSE113486 with AUC of over 94% . We identiﬁed 10 miRNAs to distinguish ovarian cancer cases from normal controls: hsa-miR-1290, hsa-miR-1233-5p, hsa-miR-1914-5p, hsa-miR-1469, hsamiR-4675, hsa-miR-1228-5p, hsa-miR-3184-5p, hsa-miR-6784-5p, hsa-mi

FAQs

Why are noninvasive biomarkers important for ovarian cancer screening?

Ovarian cancer often lacks typical early symptoms, and only a minority of patients are diagnosed at early stages. Robust, minimally invasive molecular biomarkers are needed for timely detection and treatment.

How does this study identify candidate miRNA biomarkers?

The study applies Boruta, a random-forest-based feature selection approach, to cancer and non-cancer samples from GEO (GSE106817). It then validates findings using two independent GEO datasets (GSE113486 and GSE113740).

Which machine-learning algorithms are used for classification?

Five algorithms are used: logistic regression, random forest, decision trees, artificial neural networks, and XGBoost.

Aug 5, 2026

Upload by Olivia Brown