Advancing epigenetic profiling in cervical cancer: machine learning techniques for classifying DNA methylation...

Area of Science:

• Computational Biology
• Genomics
• Machine Learning

Background:

• DNA methylation patterns are crucial biomarkers in cervical cancer.
• Predicting these epigenetic alterations can aid in diagnosis and treatment.

Purpose of the Study:

• To evaluate machine learning models for predicting DNA methylation patterns in cervical cancer.
• To compare the performance of decision-tree ensemble methods against neural networks.

Main Methods:

• Utilized decision-tree ensemble models, including Random Forest, Simple Decision Tree, and XGBoost.
• Employed neural network-based models such as Convolutional Neural Networks, Feed Forward Networks, and Wavelet Neural Networks.
• Evaluated model performance using metrics like Accuracy, Sensitivity, Specificity, and F1-score.

Main Results:

• The Random Forest model achieved a prediction accuracy of 91.35%.
• Random Forest outperformed other tree-based and neural network models in predicting DNA methylation patterns.
• Comprehensive performance evaluation confirmed the robustness of the Random Forest model.

Conclusions:

• Random Forest models demonstrate significant potential for predicting DNA methylation in cervical cancer.
• These findings suggest valuable clinical applications for improving diagnostic and treatment strategies.
• Further research into Random Forest techniques could enhance epigenetic profiling in oncology.