In order to investigate if there are geographic differences of bacterial community in the activated sludge collected from different geographic regions (eastern, northwestern, and northern parts) of China and to determine the co-occurrence patterns of bacterial community, activated sludge samples were collected from 10 municipal wastewater treatment plants located in 8 cities in China. High-throughput pyrosequencing combined with the bioinformatics analysis were used to examine the bacterial community compositions in the activated sludge samples. The result of taxonomy classifier indicated that a total of 76 genera were commonly shared by more than 7 samples, which accounted for 62 to 96 % of the classified sequences in each sample. Even though some core genera existed in all examined activated sludge samples regardless of the sampling geographic location and treatment process, significant geographic differences of bacterial community compositions among the activated sludge samples were revealed by the nonmetric multidimensional scaling (NMDS) and analyses of similarity (ANOSIM) analysis. A total of 165 pairs of significant and robust correlations (positive and negative) were identified from 61 bacterial genera based on the network analysis. The data obtained in this study could provide useful information to understand the bacterial community composition in geographically distributed wastewater treatment plants and discern the co-occurrence patterns of bacterial community.