MotivationText mining has become an important tool for biomedical research. The most fundamental text-mining task is the recognition of biomedical named entities (NER), such as genes, chemicals and diseases. Current NER methods rely on pre-defined features which try to capture the specific surface properties of entity types, properties of the typical local context, background knowledge, and linguistic information. State-of-the-art tools are entity-specific, as dictionaries and empirically optimal feature sets differ between entity types, which makes their development costly. Furthermore, features are often optimized for a specific gold standard corpus, which makes extrapolation of quality measures difficult.ResultsWe show that a completely generic method based on deep learning and statistical word embeddings [called long short-term memory network-conditional random field (LSTM-CRF)] outperforms state-of-the-art entity-specific NER tools, and often by a large margin. To this end, we compared the performance of LSTM-CRF on 33 data sets covering five different entity classes with that of best-of-class NER tools and an entity-agnostic CRF implementation. On average, F1-score of LSTM-CRF is 5% above that of the baselines, mostly due to a sharp increase in recall.Availability and implementationThe source code for LSTM-CRF is available at https://github.com/glample/tagger and the links to the corpora are available at https://corposaurus.github.io/corpora/.

Abstract. Integrated access to multiple distributed and autonomous RDF data sources is a key challenge for many semantic web applications. As a reaction to this challenge, SPARQL, the W3C Recommendation for an RDF query language, supports querying of multiple RDF graphs. However, the current standard does not provide transparent query federation, which makes query formulation hard and lengthy. Furthermore, current implementations of SPARQL load all RDF graphs mentioned in a query to the local machine. This usually incurs a large overhead in network traffic, and sometimes is simply impossible for technical or legal reasons. To overcome these problems we present DARQ, an engine for federated SPARQL queries. DARQ provides transparent query access to multiple SPARQL services, i.e., it gives the user the impression to query one single RDF graph despite the real data being distributed on the web. A service description language enables the query engine to decompose a query into sub-queries, each of which can be answered by an individual service. DARQ also uses query rewriting and cost-based query optimization to speed-up query execution. Experiments show that these optimizations significantly improve query performance even when only a very limited amount of statistical information is available. DARQ is available under GPL License at http://darq.sf.net/.

The most important way of conveying new findings in biomedical research is scientific publication. Extraction of protein–protein interactions (PPIs) reported in scientific publications is one of the core topics of text mining in the life sciences. Recently, a new class of such methods has been proposed - convolution kernels that identify PPIs using deep parses of sentences. However, comparing published results of different PPI extraction methods is impossible due to the use of different evaluation corpora, different evaluation metrics, different tuning procedures, etc. In this paper, we study whether the reported performance metrics are robust across different corpora and learning settings and whether the use of deep parsing actually leads to an increase in extraction quality. Our ultimate goal is to identify the one method that performs best in real-life scenarios, where information extraction is performed on unseen text and not on specifically prepared evaluation data. We performed a comprehensive benchmarking of nine different methods for PPI extraction that use convolution kernels on rich linguistic information. Methods were evaluated on five different public corpora using cross-validation, cross-learning, and cross-corpus evaluation. Our study confirms that kernels using dependency trees generally outperform kernels based on syntax trees. However, our study also shows that only the best kernel methods can compete with a simple rule-based approach when the evaluation prevents information leakage between training and test corpora. Our results further reveal that the F-score of many approaches drops significantly if no corpus-specific parameter optimization is applied and that methods reaching a good AUC score often perform much worse in terms of F-score. We conclude that for most kernels no sensible estimation of PPI extraction performance on new text is possible, given the current heterogeneity in evaluation data. Nevertheless, our study shows that three kernels are clearly superior to the other methods.

Time series (TS) occur in many scientific and commercial applications, ranging from earth surveillance to industry automation to the smart grids. An important type of TS analysis is classification, which can, for instance, improve energy load forecasting in smart grids by detecting the types of electronic devices based on their energy consumption profiles recorded by automatic sensors. Such sensor-driven applications are very often characterized by (a) very long TS and (b) very large TS datasets needing classification. However, current methods to time series classification (TSC) cannot cope with such data volumes at acceptable accuracy; they are either scalable but offer only inferior classification quality, or they achieve state-of-the-art classification quality but cannot scale to large data volumes. In this paper, we present WEASEL (Word ExtrAction for time SEries cLassification), a novel TSC method which is both scalable and accurate. Like other state-of-the-art TSC methods, WEASEL transforms time series into feature vectors, using a sliding-window approach, which are then analyzed through a machine learning classifier. The novelty of WEASEL lies in its specific method for deriving features, resulting in a much smaller yet much more discriminative feature set. On the popular UCR benchmark of 85 TS datasets, WEASEL is more accurate than the best current non-ensemble algorithms at orders-of-magnitude lower classification and training times, and it is almost as accurate as ensemble classifiers, whose computational complexity makes them inapplicable even for mid-size datasets. The outstanding robustness of WEASEL is also confirmed by experiments on two real smart grid datasets, where it out-of-the-box achieves almost the same accuracy as highly tuned, domain-specific methods

Circadian rhythms are essential to the temporal regulation of molecular processes in living systems and as such to life itself. Deregulation of these rhythms leads to failures in biological processes and eventually to the manifestation of pathological phenotypes including cancer. To address the questions as to what are the elicitors of a disrupted clock in cancer, we applied a systems biology approach to correlate experimental, bioinformatics and modelling data from several cell line models for colorectal and skin cancer. We found strong and weak circadian oscillators within the same type of cancer and identified a set of genes, which allows the discrimination between the two oscillator-types. Among those genes are IFNGR2, PITX2, RFWD2, PPARγ, LOXL2, Rab6 and SPARC, all involved in cancer-related pathways. Using a bioinformatics approach, we extended the core-clock network and present its interconnection to the discriminative set of genes. Interestingly, such gene signatures link the clock to oncogenic pathways like the RAS/MAPK pathway. To investigate the potential impact of the RAS/MAPK pathway - a major driver of colorectal carcinogenesis - on the circadian clock, we used a computational model which predicted that perturbation of BMAL1-mediated transcription can generate the circadian phenotypes similar to those observed in metastatic cell lines. Using an inducible RAS expression system, we show that overexpression of RAS disrupts the circadian clock and leads to an increase of the circadian period while RAS inhibition causes a shortening of period length, as predicted by our mathematical simulations. Together, our data demonstrate that perturbations induced by a single oncogene are sufficient to deregulate the mammalian circadian clock.