pubtator

is a Data Source.

PubTator 3.0 is an AI-powered biomedical literature resource that uses state-of-the-art AI techniques to provide semantic and relation searches for key biomedical entities like proteins, genetic variants, diseases and chemicals. It offers over one billion entity and relation annotations across approximately 36 million PubMed abstracts and 6 million full-text articles from the PMC open access subset, updated weekly.

Compare

Domains

literature, biomedical, health, genomics, pharmacology

Infores ID

Unknown

FAIRsharing ID

Unknown

Product Summary

Contacts

National Center for Biotechnology Information (NCBI)

Email: info@ncbi.nlm.nih.gov

URL: https://www.ncbi.nlm.nih.gov/research/pubtator3/

Publications

PubTator 3.0: an AI-powered literature resource for unlocking biomedical knowledge Preferred

Products

From this Resource

ID	Name	URL	Category	Format	Description
pubtator.api	PubTator 3.0 API	pubtator3-api	ProgrammingInterface	❔	PubTator 3.0 API for programmatic acc...
pubtator.bulk	PubTator 3.0 Bulk Downloads	PubTator3	Product	❔	Bulk downloads of annotated articles ...
pubtator.site	PubTator 3.0 Web Interface	pubtator3	GraphicalInterface	❔	Web interface for exploring PubTator ...

From other Resources

ID	Name	URL	Category	Format	Relation	Description
gnbr.graph	GNBR graph	3459420	GraphProduct	❔	had primary source	Text-mined biomedical knowledge graph...
emind.site	eMIND Web Interface	*	GraphicalInterface	http	was derived from	Web interface for browsing eMIND text...
emind.json	eMIND JSON Export	eMIND_iTextmine.json.gz (919.4 KB)	Product	json	was derived from	Gzipped JSON export of the eMIND text...

Relevant Taxa

Human (NCBITaxon:9606)

Details

PubTator 3.0

PubTator 3.0 is a comprehensive biomedical literature resource that leverages state-of-the-art AI techniques to annotate and make searchable over one billion entity and relation annotations across approximately 36 million PubMed abstracts and 6 million full-text articles from the PMC open access subset.

Features

Semantic search for six key biomedical entity types: genes, diseases, chemicals, genetic variants, species, and cell lines
Relation search across 12 common relation types between entities
Query auto-completion to enhance search accuracy
Prioritized search results based on entity relationships
Full-text search in both abstracts and open access articles
Advanced filtering options
Weekly updates to include new literature

Technical Components

PubTator 3.0 is built on several advanced NLP components, all of which are available as open-source tools:

AIONER: An all-in-one named entity recognizer for biomedical text
GNorm2: A system for gene name normalization
tmVar3: A tool for genetic variant normalization
BioREx: A comprehensive biomedical relation extraction system

These components work together to provide highly accurate entity recognition and relation extraction, making PubTator 3.0 a valuable resource for biomedical researchers and data scientists.