Deep Motif: Visualizing Genomic Sequence Classifications
Jack Lanchantin, Ritambhara Singh, Zeming Lin et al. · arXiv (Cornell University) · 2016
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.
Cited by 53 works. doi:10.48550/arxiv.1605.01133 Open-access PDF
Authors
References
- Training Very Deep Networks
- MEME SUITE: tools for motif discovery and searching
- Character-Aware Neural Language Models
- Understanding Neural Networks Through Deep Visualization
- Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning
- DNA binding sites: representation and discovery
- Quantifying similarity between motifs
- Assigning roles to DNA regulatory motifs using comparative genomics
- Modeling the specificity of protein‐DNA interactions
- SeqGL Identifies Context-Dependent Binding Signals in Genome-Wide Regulatory Element Maps