Metabolic Shifts in Marine Phytoplankton From Viral Infection and Diel Cycles Uncovered Using Dynamic Bayesian Networks

Abstract

Viruses play a fundamental role in the ecological dynamics of marine microbes by influencing host metabolic flux and altering population size via viral lysis. Despite recent advances in understanding the temporal and environmental factors that drive community-level microbial interactions, viral activity is currently unaccounted for in biogeochemical models. Moreover, the host’s metabolic response to viral infection in natural systems remains elusive. Here, we examine changes in metabolism in marine microbial communities related to virus activity and diel cycles in the surface oceans in the North Pacific Subtropical Gyre (NPSG). We use a time-varying dynamic Bayesian network (DBN) to untangle gene co-expression networks in dominant microbes associated with the environment. The resulting data suggest that viruses play a significant role in driving microbial metabolic flux on par with diel cycles. Most notably, viruses play an opportunistic role by synchronizing viral lysis with diel cycles in their photoautotrophic hosts; and continually infecting heterotrophic hosts by co-opting their alternative energy production strategies to drive replication throughout the day. This work provides a unified model that accounts for both viral activity and diel cycles to predict the distribution, variability, and trajectory of dominant microbes and their key metabolic pathways in global ocean processes.