Bacterial community (free-living vs particle-attached) assembly driven by environmental factors and a more stable network in the pre-bloom period than post-bloom

Cyanobacterial blooms are common phenomenon in freshwater ecosystems, influencing the composition and function of bacterioplankton communities. However, the drivers on the pattern and mechanism of bacterial community were still little known especially before and after bloom outbreak. Here, high-throughput sequencing of bacterial 16S rRNA gene was used to investigate the dynamics of communities with two lifestyles of free-living (FL, 0.22–3 μm) and particle-attached (PA, >3 μm) in the Lake Taihu, and co-occurrence patterns and assembly mechanism of bacteria were also explored between pre-bloom and post-bloom periods. The results exhibited that the bloom event significantly changed the composition and diversity of plankton community. Primarily, phyla of Proteobacteria, Actinobacteria and Bacteroidetes were dominated in FL bacteria, whereas Proteobacteria, Cyanobacteria and Bacteroidetes were distributed in PA bacteria. Additionally, PA bacteria had higher α-diversity than FL bacteria in two periods. Co-occurrence network analysis revealed that the network was more complex interactions in the pre-bloom than post-bloom. Null-model analysis indicated homogeneous selection was the key assembly process in the pre-bloom, while stochastic processes played predominant role in the post-bloom. Moreover, abiotic factors were significantly correlated with keystone taxa to further reveal the deterministic process in the pre-bloom (p