Tu182 - A Structured Schema to Represent Human Immunological Studies

James Overton – Knocean Inc.; Randi Vita – La Jolla Institute for Immunology; Brendan Ha – La Jolla Institute for Immunology; Anais Gambiez – La Jolla Institute for Immunology; Jason Greenbaum – La Jolla Institute for Immunology; Robert McDougal – Yale University; Gisela Gabernet – Yale University; Jeremy Gygi – Yale University; Pramod Shinde – La Jolla Institute for Immunology; Leying Guan – Yale University; Mari Kojima – La Jolla Institute for Immunology; HCC Consortium The – HIPC; Steven Kleinstein – Yale University; Bjoern Peters – La Jolla Institute for Immunology

Abstract Text: Immunological studies yield increasingly extensive and detailed datasets, making capturing and storing this data in a well-structured way, accessible for both humans and machines, essential. The Human Immunology Project Consortium (HIPC) [PMID: 23648045] has been committed to enabling the reproducibility and reuse of its findings since inception, and a dedicated Coordinating Center (HCC) is now tasked with the creation, maintenance, and expansion of a publicly accessible online resource, ImmuneSpace (immunespace.org). The ImmuneSpace platform has recently been transformed and expanded to include a centralized Knowledgebase and application programming interface (API) to promote utilization by the greater scientific community. ImmuneSpace builds on the data model developed by the Immunology Database and Analysis Portal (ImmPort) [PMID: 29485622] but implements additional standardization following the HIPC Data Standards initiative [PMID: 22343568, 26861911, 31272390, 32283555] and multiple ontology-based community data standards to support the consistent use of terminology and promote data compatibility. Upon transfer into the ImmuneSpace Knowledgebase, the data’s internal consistency is validated, and terms are standardized using uniform ontology-based labels. Study components stored in different repositories, un-parsed raw data, and computationally inaccessible study elements are curated and integrated. The study data is restructured to generalize the components found across all human immunology studies to be easily translatable into a graph-based format. In addition, ImmuneSpace provides multi-omics modules and signatures that result from systems-level profiling studies [PMID: 28842433, 36266291]. The HIPC Project and ImmuneSpace platform verify the feasibility and huge benefit of a structured approach to representing human immunological studies in deciphering system-level phenomena.