This paper discusses data commonly used to study scientific cooperation, especially over the last three decades. Among the extensive bibliography using these data, our selection of articles highlights geographical levels most used to study "Worlds" of Scientific Cooperation in order to outline the advantages and limits of such a choice. Generally, the articles examined here do not take into account changes undergone by spatial units over time, such as modifications in weight and in the centrality index. These articles also tend to focus on key scientific fields such as nanotechnology. Very few studies examine small scientific communities such as the "DNA Transcription and Repair Group", even from a sociological viewpoint. After focusing on the publishing locations of the Group's first papers, we will define the field's trajectories over time by trying to answer questions such as "How does this small community's network evolve?" "Are the original locations still central to the evolving network?" This case study of the "Worlds" of Scientific Cooperation leads to questions such as "Are scientific activities substantially different from economic activities, such as the flow of finance or goods?" "What can a multi-level approach to Scientific Cooperation offer?"