Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM 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. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.

Recurrent nets are in principle capable to store past inputs to produce the currently desired output. Because of this property recurrent nets are used in time series prediction and process control. Practical applications involve temporal dependencies spanning many time steps, e.g. between relevant inputs and desired outputs. In this case, however, gradient based learning methods take too much time. The extremely increased learning time arises because the error vanishes as it gets propagated back. In this article the de-caying error flow is theoretically analyzed. Then methods trying to overcome vanishing gradients are briefly discussed. Finally, experiments comparing conventional algorithms and alternative methods are presented. With advanced methods long time lag problems can be solved in reasonable time.

The Tox21 Data Challenge has been the largest effort of the scientific community to compare computational methods for toxicity prediction. This challenge comprised 12,000 environmental chemicals and drugs which were measured for 12 different toxic effects by specifically designed assays. We participated in this challenge to assess the performance of Deep Learning in computational toxicity prediction. Deep Learning has already revolutionized image processing, speech recognition, and language understanding but has not yet been applied to computational toxicity. Deep Learning is founded on novel algorithms and architectures for artificial neural networks together with the recent availability of very fast computers and massive datasets. It discovers multiple levels of distributed representations of the input, with higher levels representing more abstract concepts. We hypothesized that the construction of a hierarchy of chemical features gives Deep Learning the edge over other toxicity prediction methods. Furthermore, Deep Learning naturally enables multi-task learning, that is, learning of all toxic effects in one neural network and thereby learning of highly informative chemical features. In order to utilize Deep Learning for toxicity prediction, we have developed the DeepTox pipeline. First, DeepTox normalizes the chemical representations of the compounds. Then it computes a large number of chemical descriptors that are used as input to machine learning methods. In its next step, DeepTox trains models, evaluates them, and combines the best of them to ensembles. Finally, DeepTox predicts the toxicity of new compounds. In the Tox21 Data Challenge, DeepTox had the highest performance of all computational methods winning the grand challenge, the nuclear receptor panel, the stress response panel, and six single assays (teams "Bioinf@JKU"). We found that Deep Learning excelled in toxicity prediction and outperformed many other computational approaches like naive Bayes, support vector machines, and random forests.

We present primary results from the Sequencing Quality Control (SEQC) project, coordinated by the United States Food and Drug Administration. Examining Illumina HiSeq, Life Technologies SOLiD and Roche 454 platforms at multiple laboratory sites using reference RNA samples with built-in controls, we assess RNA sequencing (RNA-seq) performance for junction discovery and differential expression profiling and compare it to microarray and quantitative PCR (qPCR) data using complementary metrics. At all sequencing depths, we discover unannotated exon-exon junctions, with >80% validated by qPCR. We find that measurements of relative expression are accurate and reproducible across sites and platforms if specific filters are used. In contrast, RNA-seq and microarrays do not provide accurate absolute measurements, and gene-specific biases are observed, for these and qPCR. Measurement performance depends on the platform and data analysis pipeline, and variation is large for transcript-level profiling. The complete SEQC data sets, comprising >100 billion reads (10Tb), provide unique resources for evaluating RNA-seq analyses for clinical and regulatory settings.

The to date largest comparative study of nine state-of-the-art drug target prediction methods finds that deep learning outperforms all other competitors. The results are based on a benchmark of 1300 assays and half a million compounds.