Kingdom level microbial interaction network analysis for deciphering the potential pathobiome of oil palm leaf blight

Oil palm leaf blight is an economically significant disease but remain understudied. In this study, we hypothesized that kingdom-level microbial interactions drive the transition from asymptomatic to symptomatic states and reshape microbial ecological roles, thereby contributing to disease development. To test this, we integrated 16S rRNA gene and ITS amplicon sequencing with correlation-based network inference to examine intra- and interkingdom interactions among bacterial and fungal communities in leaf, root and soil samples. Although alpha and beta diversity analyses showed no significant differences in microbial composition between health states, network analysis revealed pronounced shifts in microbial interaction patterns. This indicates that differences in taxa connectivity, rather than taxa presence or abundance, distinguish asymptomatic from symptomatic communities. Symptomatic microbiomes exhibited more connected and restructured networks, suggesting functional reorganization. Notably, Ceramothyrium fici , one of the most abundant fungal keystone taxa in symptomatic communities, formed cross-kingdom interaction networks with bacterial genera, including Caulobacter and Streptomyces , along with other fungi, including Acremonium hennebertii , creating modules that may facilitate disease progression. Network topology further revealed increased connectivity, altered modularity and the emergence of keystone taxa. Overall, these findings suggest that oil palm leaf blight may be associated with shifts in microbial interaction networks, particularly bacterial–fungal relationships and highlight the potential value of network-based approaches for understanding disease-related microbiomes.