<i>nala</i>: text mining natural language mutation mentions

Cejuela, Juan Miguel; Bojchevski, Aleksandar; Uhlig, Carsten; Bekmukhametov, Rustem; Karn, Sanjeev Kumar; Mahmuti, Shpend; Baghudana, Ashish; Dubey, Ankit; Satagopam, Venkata; Rost, Burkhard

doi:10.1093/bioinformatics/btx083

Cited by 20 publications

(15 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To date, a handful of automatic tools have attempted to address this issue. For instance, command-line automatic variation detection tools such as EMU ( 10 ), MutationFinder ( 11 ) or Nala ( 12 ) can recognize variation mentions in text and return the results in wNm format (e.g. ‘A146T’), while SETH ( 13 ) and tmVar ( 6 ) can further map the extracted mentions to the specific dbSNP identifiers (e.g.…”

Section: Introductionmentioning

confidence: 99%

LitVar: a semantic search engine for linking genomic variant data in PubMed and PMC

Allot

Peng

Wei

et al. 2018

Nucleic Acids Research

119

View full text Add to dashboard Cite

The identification and interpretation of genomic variants play a key role in the diagnosis of genetic diseases and related research. These tasks increasingly rely on accessing relevant manually curated information from domain databases (e.g. SwissProt or ClinVar). However, due to the sheer volume of medical literature and high cost of expert curation, curated variant information in existing databases are often incomplete and out-of-date. In addition, the same genetic variant can be mentioned in publications with various names (e.g. ‘A146T’ versus ‘c.436G>A’ versus ‘rs121913527’). A search in PubMed using only one name usually cannot retrieve all relevant articles for the variant of interest. Hence, to help scientists, healthcare professionals, and database curators find the most up-to-date published variant research, we have developed LitVar for the search and retrieval of standardized variant information. In addition, LitVar uses advanced text mining techniques to compute and extract relationships between variants and other associated entities such as diseases and chemicals/drugs. LitVar is publicly available at https://www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/LitVar.

show abstract

Section: Introductionmentioning

confidence: 99%

LitVar: a semantic search engine for linking genomic variant data in PubMed and PMC

Allot

Peng

Wei

et al. 2018

Nucleic Acids Research

119

View full text Add to dashboard Cite

show abstract

“…Inversely, whereas all other variant corpora were identified to have the ‘+’ character as significant (a symbol commonly used to denote mutations), VariomeCorpus was the only not to share such characteristic. Indeed, VariomeCorpus was recently reported to annotate many vague mentions such as ‘de novo mutation’ and ‘large deletion’, with only a subset mentioning position-specific variants ( Cejuela et al , 2017 ). Due to such differences, this was excluded from subsequent power analyses.…”

Section: Resultsmentioning

confidence: 99%

Exploiting and assessing multi-source data for supervised biomedical named entity recognition

2018

View full text Add to dashboard Cite

MotivationRecognition of biomedical entities from scientific text is a critical component of natural language processing and automated information extraction platforms. Modern named entity recognition approaches rely heavily on supervised machine learning techniques, which are critically dependent on annotated training corpora. These approaches have been shown to perform well when trained and tested on the same source. However, in such scenario, the performance and evaluation of these models may be optimistic, as such models may not necessarily generalize to independent corpora, resulting in potential non-optimal entity recognition for large-scale tagging of widely diverse articles in databases such as PubMed.ResultsHere we aggregated published corpora for the recognition of biomolecular entities (such as genes, RNA, proteins, variants, drugs and metabolites), identified entity class overlap and performed leave-corpus-out cross validation strategy to test the efficiency of existing models. We demonstrate that accuracies of models trained on individual corpora decrease substantially for recognition of the same biomolecular entity classes in independent corpora. This behavior is possibly due to limited generalizability of entity-class-related features captured by individual corpora (model ‘overtraining’) which we investigated further at the orthographic level, as well as potential annotation standard differences. We show that the combined use of multi-source training corpora results in overall more generalizable models for named entity recognition, while achieving comparable individual performance. By performing learning-curve-based power analysis we further identified that performance is often not limited by the quantity of the annotated data.Availability and implementationCompiled primary and secondary sources of the aggregated corpora are available on: https://github.com/dterg/biomedical_corpora/wiki and https://bitbucket.org/iAnalytica/bioner.Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

“…The annotation of variants involves a critical step that finds literature to explain the potential influence of a discovered variant on protein functions, cell behaviors, organ normality, etc. In this regard, many efforts have been made to improve the search performance [4,5]. Even though, we observed that the sensitivity of variant queries still has space to increase.…”

Section: Introductionmentioning

confidence: 85%

variant2literature: full text literature search for genetic variants

Lin¹,

Lu²,

Chen³

et al. 2019

Preprint

View full text Add to dashboard Cite

Motivation: Whole genome sequencing (WGS) by next-generation sequencing produces millions of variants for an individual. The retrieval of biomedical literature for such a large number of genetic variants remains challenging, because in many cases the variants are only present in tables as images, or in the supplementary documents of which the file formats are diverse. Results: The proposed tool named variant2literature from the TaiGenomics (Toolkits for AI genomics) resolves the problem by incorporating text recognition with image processing. In addition to the adoption of advanced text retrieval, the recall rate of finding the literature containing the variants of interest is further improved by employing the skill of variant normalization. Different variant presentations are transformed into chromosome coordinates (standard VCF format) such that false negatives can be largely avoided. variant2literature is available in two ways. First, a web-based interface is provided to search all the literature in PMC Open Access Subset. Second, the command-line executable can be downloaded such that the users are free to search all the files in a specified directory locally.

show abstract

nala: text mining natural language mutation mentions

Cited by 20 publications

References 30 publications

LitVar: a semantic search engine for linking genomic variant data in PubMed and PMC

LitVar: a semantic search engine for linking genomic variant data in PubMed and PMC

Exploiting and assessing multi-source data for supervised biomedical named entity recognition

variant2literature: full text literature search for genetic variants

Contact Info

Product

Resources

About