DeepH&M is a deep learning-based method for simultaneously estimating absolute hydroxymethylation and methylation levels at single CpG resolution by integrating enrichment and restriction enzyme sequencing methods
DeepH&M model. (A). Schematic explanations for the 3 main assays used for DeepH&M model. (B). Structure of DeepH&M model. DeepH&M is composed of 3 modules. CpG module takes inputs of genomic features and methylation features. DNA module processes raw DNA sequence data using a convolutional neural network. Joint module combines outputs from CpG module and DNA module to predict 5hmC and 5mC simultaneously. Examples were given to show how 5hmC and 5mC were predicted from the 3 main assays. Conv is convolutional layer. Pool is pooling layer. Full con is full connected layer.
Embedded browser: gold standard 5hmC/5mC and DeepH&M predictions for 7-week-old and 79-week-old mouse cerebellum
Checkout the repository below and follow the documentation in GitHub to run DeepHM tool.
git clone https://github.com/hcharles14/DeepHM.git
Yu He, Hyo Sik Jang, Xiaoyun Xing, Daofeng Li, Michael J. Vasek, Joseph D. Dougherty, Ting Wang. DeepH&M: Simultaneous determination of single-CpG hydroxymethylation and methylation levels from enrichment and restriction enzyme sequencing methods.