Objectives: Monarch Initiative is one of the efforts to semantically integrate information across databases by employing ontologies. It is an integrative data and analytic platform connecting phenotypes to genotypes across species that imported data from 30 data sources using 24 ontologies. This study aims to understand the ontologies used in the Monarch Initiatives to account for the knowledge graph usage of the project. Knowledge graph, knowledge bases, ontologies, are interchangeably used due to lacking solid foundation on their definition, scope and most of all usage. This study also aims delineating the boundaries of these terms to better understand and use them.
Methods: Rapid evidence reviews which is a variation of systematic review is the method utilized in this study. All bio-ontologies in the Monarch Initiative are subjected for grey literature and descriptive summary of the findings is the synthesis output. Popular "grey literature" resources include clinicaltrials.gov, NIH RePORTER and the monarchinitiative.org. FINER (Feasible, Interesting, Novel, Ethical, Relevant) and PICOT (Population, Intervention, Comparison, Outcome of Interest, Time) are the 2 frameworks used for Refine Questions and Define Parameters stages. For biases identification, issues per possible rapid review process will be introduced to the study protocol and are decided to be acceptable or not given time constraints. Quality appraisal to identify the quality of the study conducted includes evidence summary matrices. Evidence synthesis includes the results of the implications for the study design and methods together with the relevance of the study.
Results: Study is still being conducted and is expected to complete by Q1 of 2021.
Conclusions: Today’s information landscape pictures knowledge graph to be a marketing-oriented term associated with the structed data representation. When Google announced a product called the Knowledge Graph in 2012, it’s considered more than an evolving project and a vision rather than a precise notion or system. It is argued to be the materialization and implementation of the Semantic Web Project. In 2013, Facebook launched its graph searched which followed by other giants like Microsoft, Amazon, Ebay etc. in using knowledge graphs. As time passed, KG has formalized specially in the field of knowledge management. It is the convergence of statistical and logical methods. Deep and machine learning are soaring to compliment the need for automation in training and testing KG data.