A cross-species knowledgebase to facilitate precision auditory neuroscience and audiology

About This Grant

Abstract A persistent challenge in hearing healthcare is that people with similar clinical hearing profiles often experience vastly different listening outcomes, likely due to hidden underlying pathologies that elude current assessments. Cross-species approaches are critical for connecting the anatomical and physiological mechanisms of disordered hearing in pre-clinical models with the real-world perceptual effects of hearing disorders in humans. Yet, the fields of auditory neuroscience and clinical audiology lack broadly accepted data standards and ontologies necessary to harmonize cross-species data and fully leverage the insights from these datasets for precision diagnostics and personalized treatments. To address this urgent gap, we propose the development of an early-stage, groundbreaking knowledgebase for cross-species precision auditory neuroscience. Our interdisciplinary team, backed by multiple federal grants, will build upon existing data resources and ongoing software development efforts to create predictive models that link sensorineural hearing loss across species, from preclinical rodent models to human subjects. This initiative will transition investigator-led resources into a community-driven, standards-based knowledgebase, following FAIR data principles to ensure accessibility, governance, and long-term usability. Initially leveraging data from our extensive cross-species research, the knowledgebase will offer comprehensive tools for comparing anatomical, neurophysiological, and perceptual data across species. Prominent colleagues from leading hearing-research institutions and companies (see letters of support) will contribute additional data and provide feedback to expand and further develop the knowledgebase, ensuring the resource scales and addresses the needs of the broader scientific community. In Specific Aim 1, we will establish rigorous data standards and ontologies to facilitate sharing diverse auditory neuroscience data, including EEG, single-unit recordings, cochlear and brain histology, non-invasive acoustic assays, surveys, and behavioral tests. Aim 2 focuses on developing standardized quantitative interfaces for cross-species comparisons, harnessing computational and statistical models to harmonize multidimensional data from various species and auditory system levels. In Aim 3, we will build a cutting-edge user interface for visualization, integrating low-code workflows and interactive large-language models, enabling even non- programmers to perform sophisticated analyses. This knowledgebase will catalyze precision diagnostics and treatments, offering transformative potential for secondary analyses, hypothesis generation, and closed-loop translational research from bench to bedside. Aligned with the NIDCD’s 2023-2027 Strategic Plan and the NIH’s Strategic Plan for Data Science, this resource is poised to drive advancements in hearing research and be sustained for long-term community impact.