Identifying Comparative Structures in Biomedical Text

Gupta, Samir; Mahmood, A. S. M. Ashique; Ross, Karen; Wu, Cathy; Vijay‐Shanker, K.

doi:10.18653/v1/w17-2326

Cited by 12 publications

(10 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The range of comparisons in biomedical literature is varied and extensive and obviously not limited to differential expression. In a previous work ( 36 ), we had developed a system that identifies and extracts information (components) from comparisons in general. While we refer to our earlier work for details of extraction of the comparison components, we briefly discuss the components of comparison here and how it relates to our task.…”

Section: Methodsmentioning

confidence: 99%

“…Comparison sentences are written in a variety of textual and syntactic forms. Despite the variations, our previously developed method ( 36 ) effectively extracts the components of these comparisons by defining patterns based on syntactic dependencies, thereby abstracting away from the variations. Example 6 consists of seven sentences that illustrate some of the variety in comparative sentences in the literature; the components of the comparisons, extracted by our system, are shown in Table 1 .…”

Section: Methodsmentioning

confidence: 99%

“…While we refer the reader to ( 36 ) for a full description of our system, we briefly describe the extraction process below and provide an example.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

DEXTER: Disease-Expression Relation Extraction from Text

et al. 2018

Self Cite

View full text Add to dashboard Cite

Gene expression levels affect biological processes and play a key role in many diseases. Characterizing expression profiles is useful for clinical research, and diagnostics and prognostics of diseases. There are currently several high-quality databases that capture gene expression information, obtained mostly from large-scale studies, such as microarray and next-generation sequencing technologies, in the context of disease. The scientific literature is another rich source of information on gene expression–disease relationships that not only have been captured from large-scale studies but have also been observed in thousands of small-scale studies. Expression information obtained from literature through manual curation can extend expression databases. While many of the existing databases include information from literature, they are limited by the time-consuming nature of manual curation and have difficulty keeping up with the explosion of publications in the biomedical field. In this work, we describe an automated text-mining tool, Disease-Expression Relation Extraction from Text (DEXTER) to extract information from literature on gene and microRNA expression in the context of disease. One of the motivations in developing DEXTER was to extend the BioXpress database, a cancer-focused gene expression database that includes data derived from large-scale experiments and manual curation of publications. The literature-based portion of BioXpress lags behind significantly compared to expression information obtained from large-scale studies and can benefit from our text-mined results. We have conducted two different evaluations to measure the accuracy of our text-mining tool and achieved average F-scores of 88.51 and 81.81% for the two evaluations, respectively. Also, to demonstrate the ability to extract rich expression information in different disease-related scenarios, we used DEXTER to extract information on differential expression information for 2024 genes in lung cancer, 115 glycosyltransferases in 62 cancers and 826 microRNA in 171 cancers. All extractions using DEXTER are integrated in the literature-based portion of BioXpress.Database URL: http://biotm.cis.udel.edu/DEXTER

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

DEXTER: Disease-Expression Relation Extraction from Text

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…To capture the up- or down-regulation expression of miRNAs in disease vs non-disease states from the research literature, we use a tool called DEXTER ( 51 ). DEXTER was designed after an extensive study of textual mentions of comparisons ( 52 ). It detects the differential expression levels as well as the location of the expression levels such as in cell lines or tissue samples, patient groups, control and others.…”

Section: System Designmentioning

confidence: 99%

emiRIT: a text-mining-based resource for microRNA information

et al. 2021

Self Cite

View full text Add to dashboard Cite

microRNAs (miRNAs) are essential gene regulators, and their dysregulation often leads to diseases. Easy access to miRNA information is crucial for interpreting generated experimental data, connecting facts across publications and developing new hypotheses built on previous knowledge. Here, we present extracting miRNA Information from Text (emiRIT), a text-miningbased resource, which presents miRNA information mined from the literature through a user-friendly interface. We collected 149 ,233 miRNA –PubMed ID pairs from Medline between January 1997 and May 2020. emiRIT currently contains ‘miRNA –gene regulation’ (69 ,152 relations), ‘miRNA disease (cancer)’ (12 ,300 relations), ‘miRNA –biological process and pathways’ (23, 390 relations) and circulatory ‘miRNAs in extracellular locations’ (3782 relations). Biological entities and their relation to miRNAs were extracted from Medline abstracts using publicly available and in-house developed text-mining tools, and the entities were normalized to facilitate querying and integration. We built a database and an interface to store and access the integrated data, respectively. We provide an up-to-date and user-friendly resource to facilitate access to comprehensive miRNA information from the literature on a large scale, enabling users to navigate through different roles of miRNA and examine them in a context specific to their information needs. To assess our resource’s information coverage, we have conducted two case studies focusing on the target and differential expression information of miRNAs in the context of cancer and a third case study to assess the usage of emiRIT in the curation of miRNA information. Database URL: https://research.bioinformatics.udel.edu/emirit/

show abstract

“…To capture the up or down-regulation expression of miRNAs in disease vs non-disease states from research literature, we use a tool called DEXTER (48). DEXTER was designed after an extensive study of textual mentions of comparisons (49). It detects the differential expression levels as well as the location of the expression levels such as in cell lines or tissue samples, patient groups, control and others.…”

Section: System Designmentioning

confidence: 99%

emiRIT: A text-mining based resource for microRNA information

Roychowdhury

Gupta

Qin

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Motivation: microRNAs (miRNAs) are essential gene regulators and their dysregulation often leads to diseases. Easy access to miRNA information is crucial for interpreting generated experimental data, connecting facts across publications, and developing new hypotheses built on previous knowledge. Here, we present emiRIT, a text mining-based resource, which presents miRNA information mined from the literature through a user-friendly interface. Results: We collected 149,233 miRNA-PubMed ID pairs from Medline between January 1997 to May 2020. emiRIT currently contains miRNA-gene regulation (60,491 relations); miRNA-disease (cancer) (12,300 relations); miRNA-biological process and pathways (23,390 relations); and circulatory miRNAs in extracellular locations (3,782 relations). Biological entities and their relation to miRNAs were extracted from Medline abstracts using publicly available and in-house developed text mining tools, and the entities were normalized to facilitate querying and integration. We built a database and an interface to store and access the integrated data, respectively. Conclusion: We provide an up-to-date and user-friendly resource to facilitate access to comprehensive miRNA information from the literature on a large-scale, enabling users to navigate through different roles of miRNA and examine them in a context specific to their information needs. To assess our resource's information coverage, in the absence of gold standards, we have conducted two case studies focusing on the target and differential expression information of miRNAs in the context of diseases. Database URL: https://research.bioinformatics.udel.edu/emirit/

show abstract

Identifying Comparative Structures in Biomedical Text

Cited by 12 publications

References 12 publications

DEXTER: Disease-Expression Relation Extraction from Text

DEXTER: Disease-Expression Relation Extraction from Text

emiRIT: a text-mining-based resource for microRNA information

emiRIT: A text-mining based resource for microRNA information

Contact Info

Product

Resources

About