The Data Hub: Enhancing Infectious Disease Research and Action with Reproducible Data Harmonization

ChallengeIn an increasingly interconnected world, with global shifts in demographics, society, and climate, the complexity of Global Health challenges is escalating. Infectious diseases, in particular, are embedded within complex socio-ecological systems that shape their emergence, spread, control, and prevention. Addressing this requires collaborative intelligence that synthesizes evidence from multiple health disciplines, complemented by contextual insights (1). However, data often remain dispersed across disciplines and siloed within domain-specific repositories and inaccessible sharepoints, which hinders efficient data collaboration. MethodsHarnessing computational sciences, we developed a flexible, self-hostable open-source software framework - the Data Hub -, supporting reproducible data harmonization and exploratory visualization for Global Health research and action. The development considered data practices and needs of potential users in research and global public health, as well as data standards relevant to the community. The software was developed using Python, PostGIS data architecture and the Django web framework, published under MIT license. ResultsThe Data Hub framework integrates a data fusion engine and an interactive dashboard for data collaboration (2). It further supports metadata management in compliance with FAIR data principles (3), enhancing data accessibility, interoperability, and reusability. With each stage of data processing and output generation encoded, the framework ensures transparency, reproducibility and verifiable workflow. A demonstration Data Hub was created using Ghana as a case study, incorporating more than 40 open data layers across 6 categories, covering diverse spatial and temporal dimensions (2). Identified areas of application include planning and conducting complex and interdisciplinary studies (e.g., cluster sampling), integrated risk assessment frameworks and analytical models (e.g., agent-based models), and data-driven simulation exercises to strengthen collaboration (e.g., epidemic investigation). ConclusionThe Data Hub represents a step toward fostering collaborative intelligence to confront the current and future multifaceted challenges in infectious disease and Global Health research. The core framework is openly accessible to interested users worldwide and undergoing testing to minimize user impact and improve usability, currently in its beta phase. Future objectives include simplifying the framework's setup process, evaluating its effectiveness in low-resource settings, and enhancing features for visual data comparison, dataset creation, and tool administration (e.g., user management, language settings). (1) Morgan OW et al. Nat Med. 2022; 28. doi:10.1038/s41591-022-01900-5.(2) Data Snack. Data Hub framework overview. 2024. Available online: datasnack.org) Wilkinson MD et al. Sci Data. 2016; 3. doi:10.1038/sdata.2016.18.