Scientist mobility, an important modern phenomenon, has been traditionally investigated by using inter-city mobility networks, allowing us to identify central cities through common network centrality measures (e.g., PageRank). However, focusing only on the structure of mobility networks does not provide a comprehensive view of the centrality of places in sustaining scientific production. Since mobility flows are constructed to simply reflect movements of people between geographical places, mobility networks may lead to underestimating the actual social diversity and value that places can extract, beyond these movements, from their inhabitants’ social networks. Indeed mobility is inherently intertwined with social ties. For example, prior collaboration with co-authors might partly explain subsequent movements to the co-authors’ cities. In this sense, the diversity of cities brought about by mobility fluxes can in fact be seen as resulting from the diversity of collaborations. Indeed, a city may thrive not simply because it attracts scientists from other cities, but also as a result of the collaborative links that the scientists living in the city have with scientists in other places. To assess and rank the importance of cities for scientific production, it is therefore essential to account for both mobility and collaboration.