The U.S. Government is investing heavily in development of surveillance systems that will detect intentionally released infectious agents in environments before they cause infection in humans or animals. Sensors capable of detecting infectious microorganisms and distinguishing them from the microbes that are ubiquitous in the environment, all in real time, are currently the focus of research efforts aimed at developing a robust surveillance system. Critical to the effectiveness of the sensor system, however, will be the integration of sensor data with multiple, relevant databases to accurately estimate the threat and predict the consequences of a biological attack. We propose a model collaborative approach to surveillance system design that combines development of novel, genomics-based pathogen sensors and a computational architecture to integrate sensor data with multiple, relevant databases to accurately estimate the threat and predict the consequences of a biological attack. The approach can also be extended to the problem of integrating multiple, heterogeneous data sources for monitoring and prediction of global infectious disease outbreaks and epidemics.