ChemEx: information extraction system for chemical data curation

Tharatipyakul, Atima; Numnark, Somrak; Wichadakul, Duangdao; Ingsriswang, Supawadee

doi:10.1186/1471-2105-13-s17-s9

Cited by 25 publications

(17 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data were collected at two different static magnetic field strengths (500 and 800 MHz) generating a series of 2 x 21 2D 1 H-15 N correlation maps measured as a function of CPMG frequency (S20 Fig). The spectra were processed using standard approaches and the program chemex [86,87] assuming a 2-state reaction [40,88], which has been applied to a few other domains [31,38]. Global fitted values of k ex and p bound were extracted (224 s -1 and 0.08) from these data.…”

Section: Methodsmentioning

confidence: 99%

A disordered encounter complex is central to the yeast Abp1p SH3 domain binding pathway

et al. 2020

View full text Add to dashboard Cite

Protein-protein interactions are involved in a wide range of cellular processes. These interactions often involve intrinsically disordered proteins (IDPs) and protein binding domains. However, the details of IDP binding pathways are hard to characterize using experimental approaches, which can rarely capture intermediate states present at low populations. SH3 domains are common protein interaction domains that typically bind proline-rich disordered segments and are involved in cell signaling, regulation, and assembly. We hypothesized, given the flexibility of SH3 binding peptides, that their binding pathways include multiple steps important for function. Molecular dynamics simulations were used to characterize the steps of binding between the yeast Abp1p SH3 domain (AbpSH3) and a proline-rich IDP, ArkA. Before binding, the N-terminal segment 1 of ArkA is pre-structured and adopts a polyproline II helix, while segment 2 of ArkA (C-terminal) adopts a 3 10 helix, but is far less structured than segment 1. As segment 2 interacts with AbpSH3, it becomes more structured, but retains flexibility even in the fully engaged state. Binding simulations reveal that ArkA enters a flexible encounter complex before forming the fully engaged bound complex. In the encounter complex, transient nonspecific hydrophobic and long-range electrostatic contacts form between ArkA and the binding surface of SH3. The encounter complex ensemble includes conformations with segment 1 in both the forward and reverse orientation, suggesting that segment 2 may play a role in stabilizing the correct binding orientation. While the encounter complex forms quickly, the slow step of binding is the transition from the disordered encounter ensemble to the fully engaged state. In this transition, ArkA makes specific contacts with AbpSH3 and buries more hydrophobic surface. Simulating the binding between ApbSH3 and ArkA provides insight into the role of encounter complex intermediates and nonnative hydrophobic interactions for other SH3 domains and IDPs in general.

show abstract

Section: Methodsmentioning

confidence: 99%

A disordered encounter complex is central to the yeast Abp1p SH3 domain binding pathway

et al. 2020

View full text Add to dashboard Cite

show abstract

“…There is growing interest in hybrid machine learning and dictionary systems such as the one described in [ 10 ], which obtains interesting performance on chemical entity recognition in patent texts. The authors of [ 55 ] use different approaches for different entity types (machine learning for chemical names, dictionary-based for organism and assay entities); given the complementary application, this is not a hybrid approach in the strict sense. A contrastive overview that also covers rule-based approaches is given in [ 56 ].…”

Section: Discussionmentioning

confidence: 99%

Entity recognition in the biomedical domain using a hybrid approach

et al. 2017

View full text Add to dashboard Cite

BackgroundThis article describes a high-recall, high-precision approach for the extraction of biomedical entities from scientific articles.MethodThe approach uses a two-stage pipeline, combining a dictionary-based entity recognizer with a machine-learning classifier. First, the OGER entity recognizer, which has a bias towards high recall, annotates the terms that appear in selected domain ontologies. Subsequently, the Distiller framework uses this information as a feature for a machine learning algorithm to select the relevant entities only. For this step, we compare two different supervised machine-learning algorithms: Conditional Random Fields and Neural Networks.ResultsIn an in-domain evaluation using the CRAFT corpus, we test the performance of the combined systems when recognizing chemicals, cell types, cellular components, biological processes, molecular functions, organisms, proteins, and biological sequences. Our best system combines dictionary-based candidate generation with Neural-Network-based filtering. It achieves an overall precision of 86% at a recall of 60% on the named entity recognition task, and a precision of 51% at a recall of 49% on the concept recognition task.ConclusionThese results are to our knowledge the best reported so far in this particular task.

show abstract

“…Due to OSRA being an open-source tool, other developers were able to implement it in their projects and to develop improvements by using it in the backend. ChemEx is a tool that combines OSRA in combination with a text mining workflow in order to mine natural-product-related data from scientific publications [ 19 ]. In 2020, the tool ChemSchematicResolver was published with Python bindings for OSRA (PyOSRA).…”

Section: Rule-based Systemsmentioning

confidence: 99%

A review of optical chemical structure recognition tools

Rajan

Brinkhaus

Zielesny³

et al. 2020

J Cheminform

View full text Add to dashboard Cite

Structural information about chemical compounds is typically conveyed as 2D images of molecular structures in scientific documents. Unfortunately, these depictions are not a machine-readable representation of the molecules. With a backlog of decades of chemical literature in printed form not properly represented in open-access databases, there is a high demand for the translation of graphical molecular depictions into machine-readable formats. This translation process is known as Optical Chemical Structure Recognition (OCSR). Today, we are looking back on nearly three decades of development in this demanding research field. Most OCSR methods follow a rule-based approach where the key step of vectorization of the depiction is followed by the interpretation of vectors and nodes as bonds and atoms. Opposed to that, some of the latest approaches are based on deep neural networks (DNN). This review provides an overview of all methods and tools that have been published in the field of OCSR. Additionally, a small benchmark study was performed with the available open-source OCSR tools in order to examine their performance.

show abstract

ChemEx: information extraction system for chemical data curation

Cited by 25 publications

References 19 publications

A disordered encounter complex is central to the yeast Abp1p SH3 domain binding pathway

A disordered encounter complex is central to the yeast Abp1p SH3 domain binding pathway

Entity recognition in the biomedical domain using a hybrid approach

A review of optical chemical structure recognition tools

Contact Info

Product

Resources

About