Automatically transforming pre- to post-composed phenotypes: EQ-lising HPO and MP

Abstract

BackgroundLarge-scale mutagenesis projects are ongoing to improve our understanding about the pathology and subsequently the treatment of diseases. Such projects do not only record the genotype but also report phenotype descriptions of the genetically modified organisms under investigation. Thus far, phenotype data is stored in species-specific databases that lack coherence and interoperability in their phenotype representations. One suggestion to overcome the lack of integration are Entity-Quality (EQ) statements. However, a reliable automated transformation of the phenotype annotations from the databases into EQ statements is still missing.ResultsHere, we report on our ongoing efforts to develop a method (called EQ-liser) for the automated generation of EQ representations from phenotype ontology concept labels. We implemented the suggested method in a prototype and applied it to a subset of Mammalian and Human Phenotype Ontology concepts. In the case of MP, we were able to identify the correct EQ representation in over 52% of structure and process phenotypes. However, applying the EQ-liser prototype to the Human Phenotype Ontology yields a correct EQ representation in only 13.3% of the investigated cases.ConclusionsWith the application of the prototype to two phenotype ontologies, we were able to identify common patterns of mistakes when generating the EQ representation. Correcting these mistakes will pave the way to a species-independent solution to automatically derive EQ representations from phenotype ontology concept labels. Furthermore, we were able to identify inconsistencies in the existing manually defined EQ representations of current phenotype ontologies. Correcting these inconsistencies will improve the quality of the manually defined EQ statements.

A new synonym-substitution method to enrich the human phenotype ontology

“…This compositional nature has provided the opportunity of logically defining the HPO terms, using the strategy known as Entity-Quality decomposition . The strategy was applied for mining skeletal phenotype descriptions from scientific literature [ 31 ] and integrating phenotype ontologies across multiple species [ 32 ]. Phenotype descriptions show high lexical variability, mainly in qualities.…”

“…Unlike the techniques described above [ 15 – 18 ], which were mainly based on the lexical properties of the ontology terms, our approach also takes the hierarchical structure of the ontology into account in order to produce synonym-substitution. Furthermore, on the basis that the HPO structure is highly compositional [ 30 – 32 ], we hypothesize that the HPO could be enriched by means of identifying those terms that include descendant terms as proper substrings. However, our method does not break down the terms into its components parts (affected entities and abnormal qualities) [ 31 , 32 ], but rather it identifies common subsequences of words shared between a term and its descendant terms.…”

“…Furthermore, on the basis that the HPO structure is highly compositional [ 30 – 32 ], we hypothesize that the HPO could be enriched by means of identifying those terms that include descendant terms as proper substrings. However, our method does not break down the terms into its components parts (affected entities and abnormal qualities) [ 31 , 32 ], but rather it identifies common subsequences of words shared between a term and its descendant terms. This makes it possible to apply the technique to the entire HPO and not restrict it to specific parts, such as musculoskeletal or skeletal phenotypic abnormalities.…”

Interoperability between phenotypes in research and healthcare terminologies—Investigating partial mappings between HPO and SNOMED CT

“…Similarly, the compositional structure of SNOMED terms has been exploited for assessing the consistency of its hierarchical structure [ 20 ]. Recent work based on the compositionality of phenotype terms investigated skeletal abnormalities [ 21 ] and clinical phenotypes across species [ 22 ]. However in the latter study, the Entity-Quality decomposition strategy yielded better results on the Mammalian Phenotype Ontology than on HPO.…”

Toward knowledge support for analysis and interpretation of complex traits

“…Several ontological approaches have been proposed to implement EQ statements [ 26 , 27 ] and, subsequently, tools have been developed to manually [ 28 ] or automatically [ 26 , 29 ] construct them. These tools rely on the existence of ontologies that define localization and trait concepts, such as the Foundational Model of Anatomy (FMA) [ 30 ] for human, or the Mouse Adult Gross Anatomy Ontology (MA) [ 31 ] for mouse, and the species-agnostic Phenotype and Trait Ontology (PATO) [ 25 ] for traits.…”