MethylCurate: Tool for Dataset Curation and Epigenetic Aging Clock Evaluation.

Area of Science:

• Epigenetics and computational biology.
• Bioinformatics and data science.

Background:

• Public DNA methylation datasets are crucial for epigenetic aging clock development.
• Existing tools face challenges in dataset retrieval and metadata harmonization from repositories like NCBI Gene Expression Omnibus (GEO).
• Inconsistent data structures and manual interventions hinder scalable clock evaluation.

Purpose of the Study:

• To develop an automated framework, MethylCurate, for efficient DNA methylation dataset processing.
• To address bottlenecks in dataset retrieval, metadata harmonization, and clock evaluation.
• To enable scalable and standardized benchmarking of epigenetic aging clocks.

Main Methods:

• Developed MethylCurate, an agentic AI framework using Python.
• Combined deterministic modules with large language model-assisted agents.
• Automated dataset retrieval, quality control, metadata extraction, harmonization, and data parsing from GEO.

Main Results:

• MethylCurate automates the retrieval and harmonization of DNA methylation datasets from GEO.
• The framework standardizes heterogeneous metadata and maps datasets to a unified format.
• Enables integrated and dialogue-driven workflow for scalable epigenetic aging clock evaluation.

Conclusions:

• MethylCurate overcomes limitations of existing resources for DNA methylation data.
• Facilitates streamlined and automated preprocessing and metadata harmonization.
• Provides a scalable platform for benchmarking epigenetic aging clocks.