LimTox: a web tool for applied text mining of adverse event and toxicity associations of compounds, drugs and genes

Cañada, Andrés; Capella-Gutierrez, Salvador; Rabal, Obdulia; Oyarzábal, Julen; Valencia, Alfonso; Krallinger, Martin

doi:10.1093/nar/gkx462

Cited by 38 publications

(21 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach provides extremely precise results, but the quantity of positive results remains modest as sentences appear in distinct forms and structure. Because of this limitation, recent approaches have incorporated methods on top of rule based extractors such as co-occurrence and machine learning systems [18,19]. We discuss the pros and cons of added methods in a later section.…”

Section: Rule Based Extractorsmentioning

confidence: 99%

“…This classifier uses a projection function called a kernel to map data onto a high dimensional space so datapoints can be easily discerned between classes [43]. This method was used to extract disease-gene associations [35,44,45], protein-protein interactions [19,46,47] and protein docking information [48]. Generally, support vector machines perform well on small datasets with large feature spaces but are slow to train as the number of datapoints becomes asymptotically large.…”

Section: Supervised Extractorsmentioning

confidence: 99%

See 1 more Smart Citation

Expanding a Database-derived Biomedical Knowledge Graph via Multi-relation Extraction from Biomedical Abstracts

Nicholson

Himmelstein

Greene

2019

Preprint

View full text Add to dashboard Cite

Knowledge graphs support multiple research efforts by providing contextual information for biomedical entities, constructing networks, and supporting the interpretation of high-throughput analyses. These databases are populated via some form of manual curation, which is difficult to scale in the context of an increasing publication rate. Data programming is a paradigm that circumvents this arduous manual process by combining databases with simple rules and heuristics written as label functions, which are programs designed to automatically annotate textual data. Unfortunately, writing a useful label function requires substantial error analysis and is a nontrivial task that takes multiple days per function. This makes populating a knowledge graph with multiple nodes and edge types practically infeasible. We sought to accelerate the label function creation process by evaluating the extent to which label functions could be re-used across multiple edge types. We used a subset of an existing knowledge graph centered on disease, compound, and gene entities to evaluate label function re-use. We determined the best label function combination by comparing a baseline database-only model with the same model but added edge-specific or edge-mismatch label functions. We confirmed that adding additional edge-specific rather than edge-mismatch label functions often improves text annotation and shows that this approach can incorporate novel edges into our source knowledge graph. We expect that continued development of this strategy has the potential to swiftly populate knowledge graphs with new discoveries, ensuring that these resources include cutting-edge results.

show abstract

Section: Rule Based Extractorsmentioning

confidence: 99%

Section: Supervised Extractorsmentioning

confidence: 99%

Expanding a Database-derived Biomedical Knowledge Graph via Multi-relation Extraction from Biomedical Abstracts

Nicholson

Himmelstein

Greene

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Information Extraction (IE) is an effective approach to summarize the knowledge into expressive forms for management and comprehension; it can be integrated with other knowledge resources for innovative discovery [1]. Examples include protein-protein interactions [2], drug-drug interaction [3], causal relationships between biological entities [4], and other topic-oriented association mining systems [5,6].…”

Section: Introductionmentioning

confidence: 99%

Context awareness and embedding for biomedical event extraction

Yan

Wong

2019

Bioinformatics

View full text Add to dashboard Cite

Motivation: Biomedical event detection is fundamental for information extraction in molecular biology and biomedical research. The detected events form the central basis for comprehensive biomedical knowledge fusion, facilitating the digestion of massive information influx from literature. Limited by the feature context, the existing event detection models are mostly applicable for a single task. A general and scalable computational model is desiderated for biomedical knowledge management. Results: We consider and propose a bottom-up detection framework to identify the events from recognized arguments. To capture the relations between the arguments, we trained a bi-directional Long Short-Term Memory (LSTM) network to model their context embedding. Leveraging the compositional attributes, we further derived the candidate samples for training event classifiers. We built our models on the datasets from BioNLP Shared Task for evaluations. Our method achieved the average F-scores of 0.81 and 0.92 on BioNLPST-BGI and BioNLPST-BB datasets respectively. Comparing with 7 state-of-the-art methods, our method nearly doubled the existing F-score performance (0.92 vs 0.56) on the BioNLPST-BB dataset. Case studies were conducted to reveal the underlying reasons.

show abstract

“…Tools such as EXTRACT (Pafilis et al, 2016) and those included in the PubTator suite (Wei, Kao and Lu, 2013) are sufficient for use in facilitating manual biocuration efforts. Furthermore, NER and semantic annotation algorithms have expanded beyond the concept types originally explored by the BioCreative challenges and now include post-translational modifications (Sun, Wang and Li, 2017), Gene Ontology terms (Ruch, 2016), metadata (Panahiazar, Dumontier and Gevaert, 2017), adverse effects (Cañada et al, 2017), and more (Tseytlin et al, 2016). Semantic annotation algorithms such as SemRep have been used to generate SemMedDB, a PubMed-scale repository of subject-predicate-object triples (Kilicoglu et.…”

Section: Introductionmentioning

confidence: 99%

Applying Citizen Science to Gene, Drug, Disease Relationship Extraction from Biomedical Abstracts

Tsueng

Nanis

Fouquier

et al. 2019

Preprint

View full text Add to dashboard Cite

Biomedical literature is growing at a rate that outpaces our ability to harness the knowledge contained therein. In order to mine valuable inferences from the large volume of literature, many researchers have turned to information extraction algorithms to harvest information in biomedical texts. Information extraction is usually accomplished via a combination of manual expert curation and computational methods. Advances in computational methods usually depends on the generation of gold standards by a limited number of expert curators. This process can be time consuming and represents an area of biomedical research that is ripe for exploration with citizen science. Citizen scientists have been previously found to be willing and capable of performing named entity recognition of disease mentions in biomedical abstracts, but it was uncertain whether or not the same could be said of relationship extraction. Relationship extraction requires training on identifying named entities as well as a deeper understanding of how different entity types can relate to one another. Here, we used the web-based application Mark2Cure ( https://mark2cure.org ) to demonstrate that citizen scientists can perform relationship extraction and confirm the importance of accurate named entity recognition on this task. We also discuss opportunities for future improvement of this system, as well as the potential synergies between citizen science, manual biocuration, and natural language processing.

show abstract

LimTox: a web tool for applied text mining of adverse event and toxicity associations of compounds, drugs and genes

Cited by 38 publications

References 16 publications

Expanding a Database-derived Biomedical Knowledge Graph via Multi-relation Extraction from Biomedical Abstracts

Expanding a Database-derived Biomedical Knowledge Graph via Multi-relation Extraction from Biomedical Abstracts

Context awareness and embedding for biomedical event extraction

Applying Citizen Science to Gene, Drug, Disease Relationship Extraction from Biomedical Abstracts

Contact Info

Product

Resources

About