RSS 2.0
Critical Assessment of Information Extraction in Biology - data sets are available from Resources/Corpora and require registration.

BioCreative IV

IAT (Task 5) Systems Descriptions [2013-07-25]

Description of Participating Systems

Below is a list of the systems that are participating in the upcoming BioCreative IV Interactive task. Please note that the systems will be ready for testing by mid August, therefore, some websites may not be yet ready with all functionalities, or may present an overview of the system. Also note that the information in the table is limited to that pertaining to the task proposed for the interactive activity.

Additional documentation on the systems and tasks is found in download section at the end of this page.

To participate in a user study involving one or more of these systems, please click here.

SystemDescriptionEntity RecognitionNormalizationCategorizationInformation ExtractionTextBrowserTarget User
ArgoAnnotation of metabolic process-related named entities, namely chemical entities and genes or gene products.gene/protein, chemicalUniProtKB (gene/protein), ChEBI (chemicals), CTD (action words)metabolic process-related action wordsAbstractFirefox, Chrome, SafariMetabolic pathway Databases, small compound databases
BioQRatorValidate retrieved articles based on relevance on protein-protein interaction informationgene/protein involved in PPIPPIAbstractFirefox, Chrome, Safari, Explorer PPI Databases
Cell FinderAnnotation of gene, expression relation and cell type in text snippets from a set of articlesgene, cell typePubMed search using MedLine Rankerexpression-related action words Text snippetsFirefox, Chrome, Safari, Explorer Databases that annotate gene expression
EGASIdentification and extraction of PPI events desribed over PubMed abstracts related to neuropatological disordersproteinPPIPPI-related action wordsFull lengthFirefox, Chrome, SafariDatabases that capture relation among proteins
MarkerRIFTriage on disease-biomarker, and extraction of these entities with normalizationgene/protein(biomarkers), diseaseEntrezGene, MeSHdisease-biomarkerdisease-biomarkerAbstractFirefox, ChromeDatabases that annotate biomarker/disease information, Pharma companies
OntogeneDetection of Gene/Chemical/Diseases and their interactionsgene/chemical/diseasesCTD identifiersGene/Chemical/Diseases interactionsAbstractFirefox, Chrome, Safari Pharma companies, disease/gene-centric databases
RLIMS-PAnnotation of kinase, substrate and site with normalization information.proteinUniProtKBphosphorylationphosphorylation-related action wordsAbstractFirefox, Chrome, Safari Phosphorylation-related databases, PPI databases
SciKnowMineTriage based on pre-trained categories of interestsystem is trained on categories of interest based on previously classified documentsFull length (PDF)Databases that need article classification in different annotation categories (phenotype, functional annotation, etc). Development of an extensible platform for other annotations and document management.
TagtogAnnotation of gene names within full-text documents especially machine-predicted documentsgeneFull lengthFirefox, ChromeModel Organism database