An ontology for Autism Spectrum Disorder (ASD) to infer ASD phenotypes from Autism Diagnostic Interview-Revised data

摘要: Display Omitted We augmented an autism ontology with SWRL rules to infer phenotypes from ADI-R items.We represented DSM diagnostic criteria for spectrum disorder in OWL.We developed a custom Protege plugin enumerating combinatorial OWL axioms.OWL reasoner thus infers autism-related questionnaire results.We evaluated the classification results data Simons Foundation. ObjectiveOur goal is create that will allow integration and reasoning subject classify subjects, based on this classification, new knowledge Autism Spectrum Disorder (ASD) related neurodevelopmental disorders (NDD). take first step toward by extending existing automatic inference of ASD Diagnostic & Statistical Manual Mental Disorders (DSM) subjects' Interview-Revised (ADI-R) assessment data. Materials methodsKnowledge regarding instruments, risk factors was added augment via Ontology Web Language class definitions semantic web rules. axioms support many-to-many relations items categories DSM. utilized whether 2642 whose obtained Foundation Research Initiative, meet DSM-IV-TR (DSM-IV) DSM-5 their ResultsWe extended adding 443 classes 632 represent phenotypes, along synonyms, environmental factors, frequency comorbidities. Applying set showed method produced accurate results: true positive negative rates inferring autistic diagnosis according DSM-IV were 1 0.065, respectively; rate 0.94. DiscussionThe allows disease high accuracy. ConclusionThe may benefit future studies serving as base ASD. In addition, NDDs, commonalities differences manifestations could be automatically inferred, contributing understanding pathophysiology.