There is a broad literature for the statistical analysis of a network as a snapshot, while the formulation of statistical frameworks for modelling populations of network data is still a developing area. Network populations are data sets where now each observation in the data comprises a network rather than a scalar quantity. In this talk I will present two modelling frameworks for making inferences for network populations. The first study provides a model-based approach for clustering network observations in a population using the Bayesian machinery. The second study provides a framework that allows inferences that explicitly capture information on cycles in a network population. Both methods will be illustrated using two real data applications in neuroscience and ecology, respectively.