GeneView: a comprehensive semantic search engine for PubMed

Abstract

Research results are primarily published in scientific literature and curation efforts cannot keep up with the rapid growth of published literature. The plethora of knowledge remains hidden in large text repositories like MEDLINE. Consequently, life scientists have to spend a great amount of time searching for specific information. The enormous ambiguity among most names of biomedical objects such as genes, chemicals and diseases often produces too large and unspecific search results. We present GeneView, a semantic search engine for biomedical knowledge. GeneView is built upon a comprehensively annotated version of PubMed abstracts and openly available PubMed Central full texts. This semi-structured representation of biomedical texts enables a number of features extending classical search engines. For instance, users may search for entities using unique database identifiers or they may rank documents by the number of specific mentions they contain. Annotation is performed by a multitude of state-of-the-art text-mining tools for recognizing mentions from 10 entity classes and for identifying protein–protein interactions. GeneView currently contains annotations for >194 million entities from 10 classes for ∼21 million citations with 271 000 full text bodies. GeneView can be searched at http://bc3.informatik.hu-berlin.de/.

Life-iNet: A Structured Network-Based Knowledge Exploration and Analytics System for Life Sciences

“…In life sciences domain, recent studies (Ernst et al, 2016;Szklarczyk et al, 2014;Thomas et al, 2012;Kim et al, 2008) rely on biomedical entity information associated with the documents to support entity-centric literature search. Most existing information retrieval systems exploit either the MeSH terms manually annotated for each PubMed article (Kim et al, 2008) or textual mentions of biomedical entities automatically recognized within the documents (Thomas et al, 2012), to capture the entity-document relatedness. Compared with traditional keyword-based systems, current entity-centric retrieval systems can identify and index entity information for documents in a more accurate way (to enable effective literature exploration), but encounter several challenges, as shown below, in supporting exploration and analysis of factual knowledge (i.e., entities and their relationships) in a given corpus.…”

Proceedings of ACL 2017, System Demonstrations

“…In life sciences domain, recent studies (Ernst et al, 2016;Szklarczyk et al, 2014;Thomas et al, 2012;Kim et al, 2008) rely on biomedical entity information associated with the documents to support entity-centric literature search. Most existing information retrieval systems exploit either the MeSH terms manually annotated for each PubMed article (Kim et al, 2008) or textual mentions of biomedical entities automatically recognized within the documents (Thomas et al, 2012), to capture the entity-document relatedness.…”

“…In life sciences domain, recent studies (Ernst et al, 2016;Szklarczyk et al, 2014;Thomas et al, 2012;Kim et al, 2008) rely on biomedical entity information associated with the documents to support entity-centric literature search. Most existing information retrieval systems exploit either the MeSH terms manually annotated for each PubMed article (Kim et al, 2008) or textual mentions of biomedical entities automatically recognized within the documents (Thomas et al, 2012), to capture the entity-document relatedness. Compared with traditional keyword-based systems, current entity-centric retrieval systems can identify and index entity information for documents in a more accurate way (to enable effective literature exploration), but encounter several challenges, as shown below, in supporting exploration and analysis of factual knowledge (i.e., entities and their relationships) in a given corpus.…”

“…An integrative view on all aspects of life sciences knowledge is still missing. Moreover, many newly emerged entities are not covered in current KBs, as the manual curation process is time-consuming and costly.• Restricted Analytic Functionality: Due to the lack of notion for factual structures, current retrieval and exploration systems have restricted functionality at analyzing entity relationshipsthey mainly focus on entity-centric literature search (Ernst et al, 2016;Thomas et al, 2012) and exploring entity co-occurrences (Kim et al, 2008). In practice, analytic functionality over factual information (e.g., drug-disease targeting prediction and distinctive disease-gene association identification) is highly desirable.…”

“…POLYSEARCH2 searches associations in a wide range of resources, including PUBMED, PUBMED CENTRAL, Wikipedia and life-science related databases. It does not support ad hoc freetext queries and does not employ dedicated recognition tools for entities or relations and always operates on the sentence level, much in contrast to SEMEDICO.GENEVIEW (Thomas et al, 2012) employs an large variety of named entity recognition tools to automatically annotate different entity classes in MEDLINE and PUBMED CENTRAL, including SNPs, species, chemicals, histone modifications, genes, protein-protein interactions (PPIs) and more. Document scoring includes field length normalization, such that term matches in titles achieve higher scores than comparable matches across a whole section.…”