Deep Motif: Visualizing Genomic Sequence Classifications

Abstract

This paper applies a deep convolutional/highway MLP framework to classify genomic sequences on the transcription factor binding site task. To make the model understandable, we propose an optimization driven strategy to extract "motifs", or symbolic patterns which visualize the positive class learned by the network. We show that our system, Deep Motif (DeMo), extracts motifs that are similar to, and in some cases outperform the current well known motifs. In addition, we find that a deeper model consisting of multiple convolutional and highway layers can outperform a single convolutional and fully connected layer in the previous state-of-the-art.

5 pages; 3 figures ; deep learning ; genomic sequence classification; understanding deep models

References in corpus (2)

• Understanding Neural Networks Through Deep Visualization
• Training Very Deep Networks

Cited by in corpus (9)

• Technical Note on Transcription Factor Motif Discovery from Importance Scores (TF-MoDISco) version 0.5.6.5
• Deep Learning for Genomics: A Concise Overview
• Fast differentiable DNA and protein sequence optimization for molecular design
• Deep Learning in Bioinformatics
• DeepHealth: Review and challenges of artificial intelligence in health informatics
• Genetic Architect: Discovering Genomic Structure with Learned Neural Architectures
• Sequential Interpretability: Methods, Applications, and Future Direction for Understanding Deep Learning Models in the Context of Sequential Data
• Unsupervised Representation Learning of DNA Sequences
• DeepFoldit -- A Deep Reinforcement Learning Neural Network Folding Proteins