A Guide to MethylationToActivity: A Deep Learning Framework That Reveals Promoter Activity Landscapes from DNA Methylomes in Individual Tumors.

Journal: Methods in molecular biology (Clifton, N.J.)
PMID: 36723810

Abstract

Genome-wide DNA methylomes have contributed greatly to tumor detection and subclassification. However, interpreting the biological impact of the DNA methylome at the individual gene level remains a challenge. MethylationToActivity (M2A) is a pipeline that uses convolutional neural networks to infer H3K4me3 and H3K27ac enrichment from DNA methylomes and thus infer promoter activity. It was shown to be highly accurate and robust in revealing promoter activity landscapes in various pediatric and adult cancers. The following will present a user-friendly guide through the model pipeline.

Authors

  • Karissa Dieseldorff Jones
    Department of Computational Biology, St. Jude Children's Research Hospital, Memphis, TN, USA.
  • Daniel Putnam
    Department of Computational Biology, St. Jude Children's Research Hospital, 262 Danny Thomas Place, Mail Stop 1135, Memphis, TN, 38105, USA.
  • Justin Williams
    Department of Computational Biology, St. Jude Children's Research Hospital, 262 Danny Thomas Place, Mail Stop 1135, Memphis, TN, 38105, USA.
  • Xiang Chen
    Yangtze Delta Region Institute (Huzhou), University of Electronic Science and Technology of China, Huzhou, Zhejiang, China.

Keywords

External Resources

Popular Topics
Recent Journals