Related Experiment Video
Updated: May 31, 2026

12:11
Methodology for Accurate Detection of Mitochondrial DNA Methylation
Published on: May 20, 2018
MethyNano: supervised contrastive pretraining enables robust and generalizable methylation detection from nanopore
Jiahui Yan1, Yujie Chen1, Yucong Gong2
1School of Control and Computer Engineering, North China Electric Power University, Beijing 102206, China.
Bioinformatics (Oxford, England)
|May 28, 2026
Summary
MethyNano, a novel deep learning framework, enhances the detection of 5-methylcytosine (5mC) in nanopore sequencing data. It improves accuracy and generalization across species and sequence contexts by integrating sequence and signal information effectively.
Area of Science:
- Epigenetics and Genomics
- Bioinformatics and Computational Biology
Background:
- 5-Methylcytosine (5mC) is crucial for gene regulation and development.
- Nanopore sequencing offers direct 5mC detection but faces challenges in generalization and signal integration.
Purpose of the Study:
- To develop an advanced deep learning framework for accurate 5mC detection from nanopore sequencing data.
- To overcome limitations of existing methods in cross-species and cross-context generalization.
Main Methods:
- Developed MethyNano, a deep learning framework utilizing a contrastive learning strategy.
- Integrated sequence and current signals for improved 5mC detection.
- Validated performance across diverse datasets (A. thaliana, O. sativa, H. sapiens).
Main Results:
- MethyNano demonstrated superior performance compared to existing methods across key metrics.
- Achieved robust generalization across species and sequence contexts (CpG/CHG/CHH).
- Showcased effective integration of sequence and signal features for higher predictive accuracy.
Conclusions:
- MethyNano offers a significant advancement in direct 5mC detection using nanopore sequencing.
- The framework provides more accurate and generalizable epigenetic profiling.
- Code availability facilitates further research and application in epigenetics.

