Simulated Infectious Disease and Pathogen Datasets for Epidemiological Modeling and Analysis

Wastewater contains rich biological signals that can be used to monitor population health, detect infectious diseases, and identify emerging outbreaks. Pathogens and other biomarkers in sewage provide a unique, noninvasive view of disease prevalence at the community scale. However, extracting these signals requires extensive field sampling, laboratory processing, and expert analysis. As a result, wastewater-based epidemiology data are scarce, geographically fragmented, and rarely available as open datasets. Even when shared, existing datasets typically span short time periods and limited regions, constraining their value for method development, benchmarking, and large-scale modeling. To address this gap, we present a framework for generating simulated infectious disease and sewer network pathogen datasets for wastewater-based epidemiology. Our approach models human mobility, disease transmission, and pathogen shedding. The resulting data provide a comprehensive, high-resolution view of the spatial and temporal dynamics of infections and wastewater measurements across an urban area. We release a fully simulated dataset comprising check-in data, social links, infection states, pathogen loads, and ground-truth disease transmission information, enabling controlled experimentation in outbreak detection, source localization, resource allocation, surveillance strategy design, and targeted public health interventions. We provide data for Fulton County, Georgia, where real wastewater measurements are available, allowing direct comparison and validation. We also demonstrate that the framework generalizes to other regions by generating data for any city with available OpenStreetMap information. By making large-scale, realistic wastewater epidemiology data readily available, this work lowers the barrier to research in outbreak monitoring, modeling, and public health decision support, and enables reproducible evaluation of methods that are otherwise infeasible with scarce real-world data.