Deep learning in epigenomics and the new Python tool MethylNet

Byte sized content for the busy scientist

Deep learning in epigenomics and the new Python tool MethylNet

Deep learning applications in everyday life

Abstract

Background

The use of deep learning in analyses of DNA methylation data is beginning to emerge and distill non-linear relationships among high-dimensional data features. However, a generalized and user-friendly approach for execution, training, and interpreting models for methylation data is lacking.

Results

We introduce and demonstrate the robust performance of MethylNet on downstream tasks of DNA methylation analysis, including cell-type deconvolution, pan-cancer classification, and subject age prediction. We interrogate the learned features from a pan-cancer classification to show high fidelity clustering of cancer subtypes, and compare the importance assigned to CpGs for the age and cell-type analyses to demonstrate concordance with expected biology.

Conclusions

Our findings demonstrate high accuracy of end-to-end deep learning methods on methylation prediction tasks. Together, our results highlight the promise of future steps to use transfer learning, hyperparameter optimization and feature interpretations on DNA methylation data.

Joshua J. Levy, Alexander J. Titus, Curtis L. Petersen, Youdinghuan Chen, Lucas A. Salas, Brock C. Christensen (2019) MethylNet: A Modular Deep Learning Approach to Methylation Prediction. bioRxiv. [DOI: 10.1101/692665] [PDF].

Read more about deep learning in biology by checking out our other articles.

 

Leave a Reply

Your email address will not be published. Required fields are marked *