Global and team science approaches are on the rise, as is attention to the network underpinnings of gender disparities in scientific collaboration. Many network studies of men’s and women’s collaboration rely on bounded case studies of single disciplines and/or single countries and limited measures related to the collaborative process. We deploy network analysis on the scholarly database Scopus to gain insight into gender inequity across regions and subject areas and to better understand contextual underpinnings of stagnancy. Using a dataset of over 1.2 million authors and 144 million collaborative relationships, we capture international and unbounded co-authorship networks that include intra- and inter-disciplinary co-authorship ties across time (2009–2013). We describe how gender informs structural features and status differences in network relationships, focusing on men and women authors in 16 region-subject pairs. We pay particular attention to how connected authors are (first- and second-order degree centrality), attributes of authors’ collaborative relationships (including the “quality” and other characteristics of these ties), tendencies towards gender homophily (proportion of same-gender ties), and the nature of men’s and women’s interdisciplinary and international reach. Men have more advantageous first-order connections, yet second-order collaborative profiles look more similar. Men and women exhibit homophilous attachment to authors of the same gender, consistent over time. There is notable variation in the level of gender disparity within subjects across countries. We discuss this variation in the context of global trends in men’s and women’s scientific participation and cultural- and country-level influences on the organization and production of science.