DeepChrome 2.0: Investigating and Improving Architectures, Visualizations, & Experiments

TL;DR

DeepChrome 2.0 advances understanding of gene regulation by introducing a novel visualization technique using GANs, comparing architectures, and demonstrating that simple models can match complex ones in predicting gene expression from histone modifications.

Contribution

It introduces a GAN-based visualization method and compares neural network architectures, showing simpler models perform as well as complex ones in gene expression prediction.

Findings

GAN-based visualization reveals combinatorial histone modification relationships.

Simple linear models perform comparably to complex CNNs.

Gene expression prediction is cell-type independent.

Abstract

Histone modifications play a critical role in gene regulation. Consequently, predicting gene expression from histone modification signals is a highly motivated problem in epigenetics. We build upon the work of DeepChrome by Singh et al. (2016), who trained classifiers that map histone modification signals to gene expression. We present a novel visualization technique for providing insight into combinatorial relationships among histone modifications for gene regulation that uses a generative adversarial network to generate histone modification signals. We also explore and compare various architectural changes, with results suggesting that the 645k-parameter convolutional neural network from DeepChrome has the same predictive power as a 12-parameter linear network. Results from cross-cell prediction experiments, where the model is trained and tested on datasets of varying sizes, cell-types, and correlations, suggest the relationship between histone modification signals and gene expression is independent of cell type. We release our PyTorch re-implementation of DeepChrome on GitHub \footnote{\url{github.com/ssss1029/gene_expression_294}}.\parfillskip=0pt