Microorganisms are the core drivers of coal biogeochemistry
and
are closely related to the formation of coalbed methane. However,
it remains poorly understood about the network relationship and stability
of microbial communities in coals with different ranks. In this study,
a high-throughput sequencing data set was analyzed to understand the
microbial co-occurrence network in coals with different ranks including
anthracite, medium-volatile bituminous, and high-volatile bituminous.
The results showed similar topological properties for the microbial
networks among coals with different ranks, but a great difference
was found in the microbial composition in different large modules
among coals with different ranks, and these three networks had three,
four, and four large modules with seven, nine, and nine phyla, respectively.
Among these networks, a total of 46 keystone taxa were identified
in large modules, and these keystone taxa were different in coals
with different ranks. Bacteria dominated the keystone taxa in the
microbial network, and these bacterial keystone taxa mainly belonged
to phyla Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria.
Besides, the removal of the key microbial data could reduce the community
stability of microbial communities in bituminous coals. A partial
least-squares path model further showed that these bacterial keystone
taxa indirectly affected methanogenic potential by maintaining the
microbial community stability and bacterial diversity. In summary,
these results showed that keystone taxa played an important role in
determining the community diversity, maintaining the microbial community
stability, and controlling the methanogenic potential, which is of
great significance for understanding the microbial ecology and the
geochemical cycle of coal seams.