Co-occurrence network and predicted functions of bacterial community in soil from Brazilian Cerrado under distinct types of vegetation cover

Abstract

The study of bacterial communities and their interactions is crucial for understanding ecosystem functioning. However, the interactions between bacterial groups and their potential functions are affected by vegetation types, mainly by more or less rich and diverse vegetation found within the Brazilian Cerrado. Thus, this study assessed the co-occurrence network and predicted functions of the soil bacterial community, comparing distinct vegetation types (open grassland, typical savanna, woodland savanna, and forest savanna) within a native Brazilian Cerrado. We observed a clear separation in the predicted functions of bacterial communities when comparing the forest savanna to other typologies. The predicted functions of the bacterial community in the forest savanna are related to organic matter degradation and nutrient cycling, while in the open grassland, these functions are associated with ammonia and sulfate oxidation. The co-occurrence network analysis revealed differences among all vegetation typologies. The co-occurrence network in the forest savanna showed a higher number of nodes (884 nodes), which represent individual bacterial taxa. In contrast, the co-occurrence networks found in the typical savanna and woodland savanna had higher numbers of edges (10,865 and 13,221 edges, respectively), i.e., connections between different bacterial taxa. The lowest number of edges (5981 edges) was observed in the open grassland. This study reinforces the influence of vegetation types on soil bacterial communities in the Cerrado, showing that specific vegetation cover influences bacterial interactions and their predicted functions.