Related Experiment Video
Updated: Aug 12, 2025
06:07
Continuous Fluorescence-Based Endonuclease-Coupled DNA Methylation Assay to Screen for DNA Methyltransferase Inhibitors
Published on: August 5, 2022
2.7K
A Guide to MethylationToActivity: A Deep Learning Framework That Reveals Promoter Activity Landscapes from DNA
Karissa Dieseldorff Jones1, Daniel Putnam1, Justin Williams2
1Department of Computational Biology, St. Jude Children's Research Hospital, Memphis, TN, USA.
Methods in Molecular Biology (Clifton, N.J.)
|February 1, 2023
Summary
MethylationToActivity (M2A) infers gene promoter activity from DNA methylomes using deep learning. This tool aids in understanding cancer biology by revealing promoter activity landscapes.
Area of Science:
- Genomics
- Epigenetics
- Computational Biology
Background:
- Genome-wide DNA methylomes are crucial for cancer detection and subtyping.
- Interpreting the biological impact of DNA methylation at individual gene promoters is challenging.
Purpose of the Study:
- To introduce MethylationToActivity (M2A), a pipeline for inferring gene promoter activity from DNA methylomes.
- To provide a user-friendly guide to the M2A model pipeline.
Main Methods:
- Utilizes convolutional neural networks (CNNs) within the M2A pipeline.
- Infers histone modifications H3K4me3 and H3K27ac enrichment from DNA methylome data.
- Predicts active gene promoter regions.
Main Results:
- M2A demonstrates high accuracy and robustness in inferring promoter activity.
- The pipeline effectively reveals promoter activity landscapes across various cancers.
- Successfully applied to both pediatric and adult cancer datasets.
Conclusions:
- M2A offers a powerful computational approach to bridge DNA methylomes and gene promoter activity.
- Facilitates a deeper understanding of cancer-specific epigenetic alterations.
- Enhances the biological interpretation of methylome data in oncology.
