Related Experiment Video
Updated: Sep 15, 2025
07:35
Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
7.6K
Use of hybrid quantum-classical algorithms for enhancing biomarker classification.
Aninda Astuti1, Pin-Keng Shih2,3, Shan-Chih Lee4
1Department of Bioinformatics and Medical Engineering, Asia University, Taichung, Taiwan.
Plos One
|July 17, 2025
Summary
Quantum machine learning (QML) can improve gene expression biomarker discovery for clear cell renal cell carcinoma (ccRCC) metastasis. Our study shows QML algorithms enhance accuracy and speed for identifying ccRCC metastasis biomarkers.
Area of Science:
- Computational Biology
- Quantum Computing
- Oncology
Background:
- Clear cell renal cell carcinoma (ccRCC) metastasis is a complex, poorly understood process contributing to high cancer lethality.
- Identifying reliable gene expression biomarkers for ccRCC metastasis is challenging for traditional machine learning methods.
- Current research lacks efficient computational approaches for analyzing complex cancer gene expression data.
Purpose of the Study:
- To investigate the application of quantum machine learning (QML) for identifying gene expression biomarkers associated with ccRCC metastasis.
- To evaluate the performance enhancement of quantum support vector classification (QSVC) and quantum neural networks (QNN) compared to classical support vector classification (SVC) and neural networks (NN).
- To assess the effectiveness of integrating the Neural Quantum Embedding (NQE) method with QSVC/QNN for improved biomarker discovery.
Main Methods:
- A two-stage approach inspired by the Neural Quantum Embedding (NQE) method was developed for binary classification tasks.
- Quantum machine learning algorithms, specifically QSVC and QNN, were employed alongside classical SVC and NN.
- Performance was evaluated based on execution time and accuracy across diverse biomedical datasets.
Main Results:
- Quantum machine learning algorithms (QSVC, QNN) demonstrated improved execution time and accuracy compared to their classical counterparts (SVC, NN).
- The integration of NQE with QSVC/QNN showed enhanced performance in identifying potential metastasis biomarkers.
- The proposed QML approach proved effective and generalizable across different biomedical datasets.
Conclusions:
- Quantum machine learning offers a promising computational framework for advancing biomarker discovery in complex diseases like ccRCC.
- The NQE-enhanced QSVC/QNN approach provides a more efficient and accurate method for identifying gene expression biomarkers related to cancer metastasis.
- This study highlights the potential of QML to overcome limitations in traditional machine learning for critical biomedical applications.
