ABSTRACT: Network methods can be useful in design, monitoring, and analysis of cluster randomized trials for control of spread of viral infections. This usefulness grows out of the role of statistical network models in molecular epidemiology and their use to design cluster randomized trials. One way to increase the use of network science in epidemiological research is to develop methods that allow mechanistic models to be expressed as statistical models. Mechanistic models (common in modeling spread of infectious disease and simulating proposed clinical trials) are grown over time by repeatedly applying a collection of stochastic rules--for example, rules governing edge formation at the individual node level. Expressing these as statistical models facilitates the use of information, including estimates of network properties and the variabilities in these estimates to simulate epidemics and the impact of interventions on epidemics. It also allows for investigation of inference on networks and the processes that operate on them, such as transmission, with or without vaccines, even in the presence of incomplete or uncertain information. As examples of statistical models, I will discuss exponential random graph models as well as the more flexible congruence class models. As illustrations, I will discuss application of these methods to a cluster randomized trial in Botswana of combination HIV prevention modalities as well as the framework for the design of a study of booster vaccine effects on forward transmission of SARS-CoV-2 variants. Motivation for the latter is driven in part by a recent recommendation, a guidance from the UK to base approval of booster vaccines with "strain changes" to target variants on results of neutralizing antibody tests and safety, but without requiring evidence of clinical efficacy.