ORIGINAL RESEARCH article
Front. Bioinform.
Sec. RNA Bioinformatics
Volume 5 - 2025 | doi: 10.3389/fbinf.2025.1571476
Tumor Tissue-of-Origin Classification Using miRNA-mRNA-lncRNA Interaction Networks and Machine Learning Methods
Provisionally accepted
- 1Department of Obstetrics and Gynecology, Institute of Clinical Medicine, University of Tartu, Tartu, Estonia
- 2Celvia CC AS, Tartu, Estonia
- 3University of Helsinki and Helsinki University Hospital, Helsinki, Finland
- 4Division of Obstetrics and Gynaecology, Department of Clinical Science, Intervention and Technology, Karolinska Institutet (KI), Huddinge, Stockholm, Sweden
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
MicroRNAs (miRNAs) are critical regulators of gene expression and play a crucial role in cancer progression through interactions with mRNA and long non-coding RNA (lncRNA). However, their potential for classifying tumors based on tissue of origin (TOO) remains underexplored. This study presents a machine learning (ML) framework that utilizes miRNAs derived from miRNA-mRNA-lncRNA networks to classify different cancer types. We employed ensemble-based classifiers, Random Forest, AdaBoost, XGBoost, and LightGBM, to achieve robust performance, with 99% classification accuracy in distinguishing 14 distinct cancer types. Recursive feature elimination (RFE) identified a minimal set of significant miRNAs, such as miR- 21-5p, miR-93-5p, and miR-10b-5p, associated with cancer-related pathways, including TGF-beta signaling, immune modulation, and epithelial-mesenchymal transition. Validation through in silico analyses, including survival analysis, drug interaction analysis, and comparisons with known biomarkers, supports the role of our framework in cancer classification and biomarker discovery for precision oncology and early cancer detection.
Keywords: miRNAs, network, machine learning, Feature Selection, Tumor tissue origin, ensemble learning
Received: 05 Feb 2025; Accepted: 14 Apr 2025.
Copyright: © 2025 Lawarde, Khatun, Lingasamy, Salumets and Modhukur. 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) or licensor 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.
* Correspondence:
Andres Salumets, Division of Obstetrics and Gynaecology, Department of Clinical Science, Intervention and Technology, Karolinska Institutet (KI), Huddinge, 141 86, Stockholm, Sweden
Vijayachitra Modhukur, Department of Obstetrics and Gynecology, Institute of Clinical Medicine, University of Tartu, Tartu, Estonia
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.