Many real-world networks have associated metadata that assigns categorical labels to nodes. Analysis of these annotations can complement the topological analysis of complex networks. Annotated networks have typically been used to evaluate community detection approaches. Here, we introduce an approach that combines the quantitative analysis of annotations and network structure, which groups nodes according to similar distributions of node annotations in their neighbourhoods. Importantly the nodes that are grouped together, which we call homologues may not be connected to each other at all. By applying our approach to three very different real-world networks we show that these groupings identify common functional roles and properties of nodes in the network.