Bidirectional Long Short-Term Memory Networks for Predicting the Subcellular Localization of Eukaryotic Proteins
Issue No. 03 - July-September (2007 vol. 4)
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 ranged 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% novel non-plant proteins and 88.4% novel plant proteins correctly, which is an improvement over feedforward and standard recurrent networks solving the same problem. The BLSTM system is available as a web-service (http://www.stepc.gr/~synaptic/blstm.html).
recurrent neural networks, long shortterm memory, biological sequence analysis, protein subcellular localization prediction
M. Reczko and T. Thireou, "Bidirectional Long Short-Term Memory Networks for Predicting the Subcellular Localization of Eukaryotic Proteins," in IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 4, no. , pp. 441-446, 2007.