Related Experiment Video
Updated: Aug 16, 2025

07:41
A Robust Discovery Platform for the Identification of Novel Mediators of Melanoma Metastasis
Published on: March 8, 2022
2.5K
Biomarker Discovery for Meta-Classification of Melanoma Metastatic Progression Using Transfer Learning.
Jose Marie Antonio Miñoza1, Jonathan Adam Rico2, Pia Regina Fatima Zamora2
1System Modeling and Simulation Laboratory, Department of Computer Science, University of the Philippines Diliman, Quezon City 1101, Philippines.
Genes
|December 23, 2022
Summary
This study introduces a novel transfer learning model for identifying melanoma biomarkers, improving diagnosis and prognosis. The model achieved high accuracy, discovering key genes for classification and predicting patient outcomes.
Area of Science:
- Oncology
- Genomics
- Bioinformatics
Background:
- Melanoma is an aggressive skin cancer where metastasis significantly impacts prognosis.
- Accurate diagnosis and prognosis are crucial for effective melanoma treatment.
- Identifying reliable biomarkers is essential for improving patient outcomes.
Purpose of the Study:
- To develop a transfer learning-based model for melanoma biomarker discovery.
- To aid in the diagnosis and prognosis of melanoma.
- To identify novel diagnostic and prognostic gene biomarkers.
Main Methods:
- Developed a transfer learning-based biomarker discovery model.
- Applied the model to an ensemble machine learning framework.
- Validated the model using an independent dataset.
Main Results:
- The model identified known melanoma biomarkers and discovered novel ones.
- Achieved high performance with an AUC of 0.9861, 91.05% accuracy, and 90.60% F1 score on validation data.
- Identified specific genes (e.g., C7, GRIK5, S100A7, KRT14) for diagnostic and prognostic applications.
Conclusions:
- Transfer learning is a valuable approach for melanoma biomarker discovery.
- The identified genes show potential for clinical application in melanoma diagnosis and prognosis.
- The developed model can enhance the understanding and management of melanoma.

