Systematic benchmarking of tools for CpG methylation detection from nanopore sequencing

Area of Science:

• Genomics
• Epigenetics
• Bioinformatics

Background:

• DNA methylation is crucial for gene expression and genome stability.
• Accurate detection of DNA methylation from Nanopore sequencing is essential but challenging.
• Existing tools for Nanopore methylation detection lack comprehensive accuracy assessment.

Purpose of the Study:

• To systematically benchmark the accuracy of DNA methylation detection tools using Nanopore sequencing data.
• To identify strategies for improving methylation detection accuracy.
• To introduce METEORE, a consensus-based approach for enhanced methylation calling.

Main Methods:

• Benchmarking of multiple CpG methylation detection tools on Nanopore sequencing data.
• Utilized individual reads, controlled mixtures, and bisulfite sequencing for validation.
• Developed and evaluated METEORE, a consensus strategy combining multiple tool predictions.
• Provided Snakemake pipelines for reproducibility.

Main Results:

• Individual tools exhibit a trade-off between false positives and false negatives.
• Significant dispersion in methylation frequency predictions was observed across tools.
• The METEORE consensus approach demonstrated improved accuracy compared to individual tools.
• Benchmarking identified key performance variations among different methylation callers.

Conclusions:

• Current Nanopore sequencing methylation detection tools require careful evaluation due to variable accuracy.
• A consensus approach like METEORE can significantly enhance the reliability of CpG methylation detection.
• Standardized benchmarking and reproducible pipelines are vital for advancing epigenomic analysis.