DNA methylation and machine learning: challenges and perspective toward enhanced clinical diagnostics

Area of Science:

• Epigenetics and Bioinformatics
• Artificial Intelligence in Medicine

Background:

• DNA methylation, an epigenetic process, regulates gene expression and is crucial in disease development.
• Advances in high-throughput sequencing and array technologies have generated extensive biological data.
• Machine learning (ML) methods are increasingly used to analyze complex biological data for medical applications.

Purpose of the Study:

• To review recent advancements in DNA methylation research utilizing machine learning.
• To highlight the application of ML in developing precise, comprehensive, and rapid patient diagnostics.
• To provide a general workflow for researchers applying ML to DNA methylation data.

Main Methods:

• Review of recent literature on DNA methylation and machine learning applications.
• Analysis of bioinformatics data generated from arrays and sequencing.
• Showcasing successful diagnostic applications in various diseases.

Main Results:

• Machine learning significantly improves the precision and speed of diagnostics based on DNA methylation markers.
• Successful examples include the diagnosis of cancer, neurodevelopmental disorders, and multifactorial diseases.
• Some ML-driven diagnostic platforms have successfully entered the healthcare market.

Conclusions:

• Machine learning offers a powerful approach to leverage DNA methylation data for advanced diagnostics.
• The integration of ML with epigenetic studies holds significant promise for the future of personalized medicine.
• This field is rapidly evolving, with demonstrated clinical utility and market adoption.