Modeling Epidemics in Seed Systems: The Case of Sweetpotato in Northern Uganda

Abstract Seed systems are critical for deployment of improved varieties, and serve as major conduits for the spread of seed-borne pathogens. Vegetatively propagated crops in low-income countries are particularly vulnerable to seed degeneration, where yield is lowered through successive cycles of propagation because of pathogen accumulation. We evaluated the structure of an informal sweetpotato seed system for its vulnerability to the spread of epidemics, and its utility for disseminating improved varieties. During the 2014 growing season (April-Oct), vine sellers were surveyed weekly in the Gulu Region of Northern Uganda. Our analysis draws on tools from network theory to evaluate the potential for epidemic spread in this region. Utilizing empirical 2014 seed transaction data and estimated spatial spread, we constructed a seed transaction network, which was used to simulate the introduction of a pathogen, and evaluated the influence of both epidemic starting point and quarantine treatments on epidemic progress. Results indicate that the starting position in the network is critical for epidemic progress and final epidemic outcomes. Quarantine of 30 villages lowered epidemic progress up to 65.7%, when compared to the control (no quarantine), over 20 timesteps in 500 realizations. The percent control conferred by quarantine treatments was also influenced by the epidemic starting point. Considering equal likelihood of any node being an introduction point for a new epidemic, villages of particular utility for disease monitoring were also identified. Sensitivity analysis identified important parameters and priorities for future data collection. We compared the utility of node degree, betweenness, closeness, and eigenvector centrality for selecting quarantine locations, finding that betweenness had more limited utility. This analysis pipeline can be applied to provide recommendations for a wide variety of seed systems.