A Lexical Approach to Identifying Subtype Inconsistencies in Biomedical Terminologies

Abeysinghe, Rashmie; Zheng, Fengbo; Hinderer, Eugene W.; Moseley, Hunter N. B.; Cui, Licong

doi:10.1109/bibm.2018.8621511

Cited by 7 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In prior work, we introduced a similar approach leveraging word differences between concepts to audit the Gene Ontology [35,36]. However, the criteria for the selection of concept-pairs were different than what was used in this work.…”

Section: Comparison With Related Workmentioning

confidence: 99%

“…Non-lattice subgraphs indicate ontology fragments that violate lattice-property, a desirable structural indicator for a well-formed ontology [34]. Additionally, we have investigated a lexical-based inference approach to explore lexical irregularities between GO concept-pairs with and without is-a relations [35,36]. To our knowledge, such systematic approaches targeted to auditing VO have not been studied in prior work.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of missing hierarchical relations in the vaccine ontology using acquired term pairs

Manuel

Abeysinghe

et al. 2022

J Biomed Semant

Self Cite

View full text Add to dashboard Cite

Background The Vaccine Ontology (VO) is a biomedical ontology that standardizes vaccine annotation. Errors in VO will affect a multitude of applications that it is being used in. Quality assurance of VO is imperative to ensure that it provides accurate domain knowledge to these downstream tasks. Manual review to identify and fix quality issues (such as missing hierarchical is-a relations) is challenging given the complexity of the ontology. Automated approaches are highly desirable to facilitate the quality assurance of VO. Methods We developed an automated lexical approach that identifies potentially missing is-a relations in VO. First, we construct two types of VO concept-pairs: (1) linked; and (2) unlinked. Each concept-pair further derives an Acquired Term Pair (ATP) based on their lexical features. If the same ATP is obtained by a linked concept-pair and an unlinked concept-pair, this is considered to indicate a potentially missing is-a relation between the unlinked pair of concepts. Results Applying this approach on the 1.1.192 version of VO, we were able to identify 232 potentially missing is-a relations. A manual review by a VO domain expert on a random sample of 70 potentially missing is-a relations revealed that 65 of the cases were valid missing is-a relations in VO (a precision of 92.86%). Conclusions The results indicate that our approach is highly effective in identifying missing is-a relation in VO.

show abstract

Section: Comparison With Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Identification of missing hierarchical relations in the vaccine ontology using acquired term pairs

Manuel

Abeysinghe

et al. 2022

J Biomed Semant

Self Cite

View full text Add to dashboard Cite

show abstract

“…Despite maintenance and standard policies for adding terms, ontological organization is still subject to human error and disagreement, necessitating quality assurance and revising, especially as ontologies evolve or merge. A recent review of current methods for biomedical ontology mapping highlights the importance in developing semi-automatic methods [18,19] to aid in ontology evolution efforts and reiterates the aforementioned concept of semantic correspondence in terms of scoping between terms [20]. Methods incorporating such correspondences have been published elsewhere, but these deal with issues of ontology evolution and merging, and not with categorizing terms into user-defined subsets [21,22].…”

Section: Maintenance Of Ontologiesmentioning

confidence: 99%

GOcats: A tool for categorizing Gene Ontology into subgraphs of user-defined concepts

Hinderer

Moseley

2020

PLoS ONE

Self Cite

View full text Add to dashboard Cite

Gene Ontology is used extensively in scientific knowledgebases and repositories to organize a wealth of biological information. However, interpreting annotations derived from differential gene lists is often difficult without manually sorting into higher-order categories. To address these issues, we present GOcats, a novel tool that organizes the Gene Ontology (GO) into subgraphs representing user-defined concepts, while ensuring that all appropriate relations are congruent with respect to scoping semantics. We tested GOcats performance using subcellular location categories to mine annotations from GO-utilizing knowledgebases and evaluated their accuracy against immunohistochemistry datasets in the Human Protein Atlas (HPA). In comparison to term categorizations generated from UniProt's controlled vocabulary and from GO slims via OWLTools' Map2Slim, GOcats outperformed these methods in its ability to mimic human-categorized GO term sets. Unlike the other methods, GOcats relies only on an input of basic keywords from the user (e.g. biologist), not a manually compiled or static set of top-level GO terms. Additionally, by identifying and properly defining relations with respect to semantic scope, GOcats can utilize the traditionally problematic relation, has_part, without encountering erroneous term mapping. We applied GOcats in the comparison of HPA-sourced knowledgebase annotations to experimentallyderived annotations provided by HPA directly. During the comparison, GOcats improved correspondence between the annotation sources by adjusting semantic granularity. GOcats enables the creation of custom, GO slim-like filters to map fine-grained gene annotations from gene annotation files to general subcellular compartments without needing to handselect a set of GO terms for categorization. Moreover, GOcats can customize the level of semantic specificity for annotation categories. Furthermore, GOcats enables a safe and more comprehensive semantic scoping utilization of go-core, allowing for a more complete utilization of information available in GO. Together, these improvements can impact a variety of GO knowledgebase data mining use-cases as well as knowledgebase curation and quality control.

show abstract

“…Quesada-Martínez et al analyzed concept names in the SNOMED CT to identify lexical regularities and suggested missing relations [11]. Abeysinghe et al introduced a lexical-based inference approach to derive hierarchical inconsistencies and uncover missing IS-A relations in the SNOMED CT, NCI Thesaurus and Gene Ontology [12]. Liu et al created embeddings for each concept based on its related IS-A relations and used convolutional neural network to discover missing IS-A relations between neoplasm concepts in the NCI Thesaurus [13].…”

Section: Related Work On Identifying Missing Is-a Relationsmentioning

confidence: 99%

Detecting missing IS-A relations in the NCI Thesaurus using an enhanced hybrid approach

Zheng

Abeysinghe

Sioutos

et al. 2020

BMC Med Inform Decis Mak

Self Cite

View full text Add to dashboard Cite

Background The National Cancer Institute (NCI) Thesaurus provides reference terminology for NCI and other systems. Previously, we proposed a hybrid prototype utilizing lexical features and role definitions of concepts in non-lattice subgraphs to identify missing IS-A relations in the NCI Thesaurus. However, no domain expert evaluation was provided in our previous work. In this paper, we further enhance the hybrid approach by leveraging a novel lexical feature—roots of noun chunks within concept names. Formal evaluation of our enhanced approach is also performed. Method We first compute all the non-lattice subgraphs in the NCI Thesaurus. We model each concept using its role definitions, words and roots of noun chunks within its concept name and its ancestor’s names. Then we perform subsumption testing for candidate concept pairs in the non-lattice subgraphs to automatically detect potentially missing IS-A relations. Domain experts evaluated the validity of these relations. Results We applied our approach to 19.08d version of the NCI Thesaurus. A total of 55 potentially missing IS-A relations were identified by our approach and reviewed by domain experts. 29 out of 55 were confirmed as valid by domain experts and have been incorporated in the newer versions of the NCI Thesaurus. 7 out of 55 further revealed incorrect existing IS-A relations in the NCI Thesaurus. Conclusions The results showed that our hybrid approach by leveraging lexical features and role definitions is effective in identifying potentially missing IS-A relations in the NCI Thesaurus.

show abstract

A Lexical Approach to Identifying Subtype Inconsistencies in Biomedical Terminologies

Cited by 7 publications

References 18 publications

Identification of missing hierarchical relations in the vaccine ontology using acquired term pairs

Identification of missing hierarchical relations in the vaccine ontology using acquired term pairs

GOcats: A tool for categorizing Gene Ontology into subgraphs of user-defined concepts

Detecting missing IS-A relations in the NCI Thesaurus using an enhanced hybrid approach

Contact Info

Product

Resources

About