Deep Motif Dashboard: Visualizing and Understanding Genomic Sequences Using Deep Neural Networks

By Jack Lanchantin, Ritambhara Singh, Beilun Wang, Yanjun Qi. 2016

136 citations recorded by the source corpus.

Deep neural network (DNN) models have recently obtained state-of-the-art prediction accuracy for the transcription factor binding (TFBS) site classification task. However, it remains unclear how these approaches identify meaningful DNA sequence signals and give insights as to why TFs bind to certain locations. In this paper, we propose a toolkit called the Deep Motif Dashboard (DeMo Dashboard) which provides a suite of visualization strategies to extract motifs, or sequence patterns from deep neural network models for TFBS classification. We demonstrate how to visualize and understand three important DNN models: convolutional, recurrent, and convolutional-recurrent networks. Our first visualization method is finding a test sequence’s saliency map which uses first-order derivatives to describe the importance of each nucleotide in making the final prediction. Second, considering recurrent models make predictions in a temporal manner (from one end of a TFBS sequence to the other), we introduce temporal output scores, indicating the prediction score of a model over time for a sequential input. Lastly, a class-specific visualization strategy finds the optimal input sequence for a given TFBS positive class via stochastic gradient optimization. Our experimental results indicate that a convolutional-recurrent architecture performs the best among the three architectures. The visualization techniques indicate that CNN-RNN makes predictions by modeling both motifs as well as dependencies among them.

Subfields: Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Opportunities and obstacles for deep learning in biology and medicine (2017, methodology): “[165] looks at how the output of a recurrent neural network changes as longer and longer subsequences are supplied as input to the network, where the subsequences begin with just the first position and end with the entire sequence.”
Biological Sequence Modeling with Convolutional Kernel Networks (2017, neutral): “However, a systematic approach is still lacking as existing methods rely on speciﬁc sequences to interpret trained ﬁlters [Alipanahi et al., 2015, Shrikumar et al., 2017a] or output a single feature per class [Lanchantin et al., 2017, (3.3)].”
Attend and Predict: Understanding Gene Regulation by Selective Attention on Chromatin (2017, background): “For instance, multiple recent studies have successfully used deep learning for modeling protein sequences [23, 37] and DNA sequences [1, 20], predicting gene expressions [29], as well as understanding the effects of non-coding variants [38, 27].”
Maximum entropy methods for extracting the learned features of deep neural networks (2017, background): “Methods for extracting learned input features from these networks can provide valuable scientific insights, and several efforts in this direction [6, 8] have made deep learning an even more appealing approach for tackling complex problems in genomics and other scientific disciplines.”
Obtaining genetics insights from deep learning via explainable artificial intelligence (2022, background): “No consensus currently exists regarding how to design the best neural network architecture for a given task, so researchers typically experiment with multiple architectures 21 .”