Biogeographic patterns of microbial co-occurrence ecological networks in six American forests

Rather than simple accumulation of individual populations, microorganisms in natural ecosystems form complex ecological networks that are critical to maintain ecosystem functions and services. Although various studies have examined the patterns of microbial community diversity and composition across spatial gradients, whether microbial co-occurrence relationships follow similar patterns remains an open question. In this study, we determined the biogeographic patterns of microbial co-occurrence networks of bacteria, fungi and nitrogen (N) fixer via analyses of high throughput amplicon sequencing data of 16S rRNA, ITS, and nifH genes from 126 forest soil samples across six forests in America. Microbial co-occurrence networks were constructed using a Random Matrix Theory based approach. Network parameters were calculated and correlated with biogeographic parameters. Gradient patterns along with biogeographic parameters were observed for network topologies. Significantly different network topologies were observed between microbial co-occurrence networks in tropical and temperate forest ecosystems. Among various biogeographic parameters potentially related with network topology indices, temperature seemed to be the strongest one. These results suggest that biogeographic variables like temperature not only mediate microbial community diversity and composition, but also the co-occurrence ecological networks among microbial species.