Bidirectional Long Short-Term Memory Networks for Predicting the Subcellular Localization of Eukaryotic Proteins

@article{Thireou2007BidirectionalLS,
  title={Bidirectional Long Short-Term Memory Networks for Predicting the Subcellular Localization of Eukaryotic Proteins},
  author={Trias Thireou and Martin Reczko},
  journal={IEEE/ACM Transactions on Computational Biology and Bioinformatics},
  year={2007},
  volume={4}
}
  • T. Thireou, M. Reczko
  • Published 1 July 2007
  • Computer Science
  • IEEE/ACM Transactions on Computational Biology and Bioinformatics
An algorithm called bidirectional long short-term memory networks (BLSTM) for processing sequential data is introduced. This supervised learning method trains a special recurrent neural network to use very long-range symmetric sequence context using a combination of nonlinear processing elements and linear feedback loops for storing long-range context. The algorithm is applied to the sequence-based prediction of protein localization and predicts 93.3 percent novel nonplant proteins and 88.4… 

Figures and Tables from this paper

Identifying Protein-Protein Interaction Using Tree LSTM and Structured Attention
TLDR
This paper proposes a novel tree recurrent neural network with structured attention architecture for doing PPI that achieves state of the art results (precision, recall, and F1-score) on the AIMed and BioInfer benchmark data sets.
Classification of Antibacterial Peptides Using Long Short-Term Memory Recurrent Neural Networks
TLDR
This work demonstrates the application of Long Short-Term Memory recurrent neural networks to classification of antibacterial peptides and compares it to a Random Forest classifier and a k-nearest neighbor classifier.
A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures
TLDR
The LSTM cell and its variants are reviewed and their variants are explored to explore the learning capacity of the LSTm cell and the L STM networks are divided into two broad categories:LSTM-dominated networks and integrated LSTS networks.
Supervised Sequence Labelling with Recurrent Neural Networks
  • A. Graves
  • Computer Science
    Studies in Computational Intelligence
  • 2008
TLDR
A new type of output layer that allows recurrent networks to be trained directly for sequence labelling tasks where the alignment between the inputs and the labels is unknown, and an extension of the long short-term memory network architecture to multidimensional data, such as images and video sequences.
USPNet: unbiased organism-agnostic signal peptide predictor with deep protein language model
TLDR
Unbiased Organism-agnostic Signal Peptide Network (USPNet) is presented, a signal peptide prediction and cleavage site prediction model based on deep protein language model that uses label distribution-aware margin (LDAM) loss and evolutionary scale modeling (ESM) embedding to handle data imbalance and object-dependence problems.
Detecting Novel Sequence Signals in Targeting Peptides Using Deep Learning
TLDR
This work presents TargetP 2.0, a novel state of art method to identify N-terminal sorting signals, which direct proteins to the secretory pathway, mitochondria and chloroplasts or other plastids, by examining the strongest signals from the attention layer in the network and finds that the second residue in the protein, i.e. the one following the initial methionine, has a strong influence on the classification.
Analyzing protein dynamics from fluorescence intensity traces using unsupervised deep learning network
TLDR
An unsupervised deep learning network approach to analyze the dynamics of membrane proteins from the fluorescence intensity traces that facilitates training of the system without predefined state number or pre-labelling and can even extract information from noise distribution.
Detecting sequence signals in targeting peptides using deep learning
During the development of TargetP 2.0, a state-of-the-art method to predict targeting signal, we find a previously overlooked biological signal for subcellular targeting using the output from a deep
Deep neural networks for human microRNA precursor detection
TLDR
Deep neural networks (DNN) could be utilized for the human pre-miRNAs detection with high performance and through proper regularization, the deep learning models, although trained on comparatively small dataset, had strong generalization ability.
Deep Bi-directional Long Short-Term Memory Neural Networks for Sentiment Analysis of Social Data
TLDR
Deep Bi-directional Long Short-Term Memory architecture with multi-levels feature presentation for sentiment polarity classification (SPC) on social data is proposed and the performance of the model is competitive with state-of-the-art of SPC on Twitter’s data.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 29 REFERENCES
Prediction of the subcellular localization of eukaryotic proteins using sequence signals and composition
TLDR
A tool based on bidirectional recurrent neural networks trained to read sequentially the amino acid sequence and produce localization information along the sequence leads to a 91% correct localization prediction for novel proteins in fivefold cross‐validation.
Finding Signal Peptides in Human Protein Sequences Using Recurrent Neural Networks
TLDR
A new approach called Sigfind for the prediction of signal peptides in human protein sequences is introduced, based on the bidirectional recurrent neural network architecture and a better learning algorithm, which results in a very accurate identification of sign peptides.
Using neural networks for prediction of the subcellular location of proteins.
TLDR
With the subcellular location restricting a protein's possible function, this method should be a useful tool for the systematic analysis of genome data and is available via a server on the world wide web.
Extensive feature detection of N-terminal protein sorting signals
TLDR
This work has succeeded in finding rules whose prediction accuracies come close to that of TargetP, while still retaining a very simple and interpretable form.
Learning Precise Timing with LSTM Recurrent Networks
TLDR
This work finds that LSTM augmented by "peephole connections" from its internal cells to its multiplicative gates can learn the fine distinction between sequences of spikes spaced either 50 or 49 time steps apart without the help of any short training exemplars.
Long Short-Term Memory
TLDR
A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Adaptive encoding neural networks for the recognition of human signal peptide cleavage sites
TLDR
An adaptive encoding artificial neural network (ACN) for recognition of sequence patterns is described and knowledge of physico-chemical properties is not necessary for training an ACN.
Support vector machine approach for protein subcellular localization prediction
TLDR
Support Vector Machine has been introduced to predict the subcellular localization of proteins from their amino acid compositions and can be a complementary method to other existing methods based on sorting signals.
Machine learning approaches for the prediction of signal peptides and other protein sorting signals.
TLDR
A hidden Markov model version of SignalP has been developed, making it possible to discriminate between cleaved signal peptides and uncleaved signal anchors, and it is shown how SignalP can be used to characterize putative signal peptide from an archaeon, Methanococcus jannaschii.
Using subsite coupling to predict signal peptides.
  • K. Chou
  • Biology
    Protein engineering
  • 2001
TLDR
Based on a model that takes into account the coupling effect among some key subsites, the so-called [-3, -1, +1] coupling model, a new prediction algorithm is developed that accuracy for secretory proteins and 1440 non-secretary proteins was over 92%.
...
1
2
3
...