Protein Secondary Structure Prediction with Long Short Term Memory Networks

Abstract

Prediction of protein secondary structure from the amino acid sequence is a classical bioinformatics problem. Common methods use feed forward neural networks or SVMs combined with a sliding window, as these models does not naturally handle sequential data. Recurrent neural networks are an generalization of the feed forward neural network that naturally handle sequential data. We use a bidirectional recurrent neural network with long short term memory cells for prediction of secondary structure and evaluate using the CB513 dataset. On the secondary structure 8-class problem we report better performance (0.674) than state of the art (0.664). Our model includes feed forward networks between the long short term memory cells, a path that can be further explored.

v2: adds larger network with slightly better results, update author affiliations

References in corpus (5)

• ADADELTA: An Adaptive Learning Rate Method
• Theano: new features and speed improvements
• Learning to Execute
• How to Construct Deep Recurrent Neural Networks
• Learning to Discover Efficient Mathematical Identities

Cited by in corpus (10)

• LSTM: A Search Space Odyssey
• Computational Protein Design with Deep Learning Neural Networks
• High Quality Prediction of Protein Q8 Secondary Structure by Diverse Neural Network Architectures
• Deep Learning in Bioinformatics
• Protein Secondary Structure Prediction Using Deep Multi-scale Convolutional Neural Networks and Next-Step Conditioning
• Next-Step Conditioned Deep Convolutional Neural Networks Improve Protein Secondary Structure Prediction
• MUST-CNN: A Multilayer Shift-and-Stitch Deep Convolutional Architecture for Sequence-based Protein Structure Prediction
• Machine Learning for Protein Function
• Machine Learning for Classification of Protein Helix Capping Motifs
• GL-BLSTM: a novel structure of bidirectional long-short term memory for disulfide bonding state prediction