Interpreting phage ecology, theory, and models in the genomic age

Dinesh Kumar Kuppa Baskaran , Neha Kashyap, James C Kosmopoulos, Marguerite V Langwig, Kris Sankaran, Karthik Anantharaman

Viruses, particularly bacteriophages, are the most abundant biological entities across nearly all ecosystems and play a central role in shaping microbial community structure, ecosystem function, and evolution. Consequently, there has been growing interest in studying phages and their interactions within microbiomes. Mathematical modeling has long provided a foundation for investigating phage–host interactions, formalizing diverse infection strategies (lytic, lysogenic and chronic infection) and their governing parameters. These approaches include resource-based trophic models that describe the flow of resources from microbes to viruses, epidemiological models that focus on the transmission of infection, stochastic models that capture environmental and demographic randomness, metacommunity models that account for spatial organization, and network-based models that characterize the topology of phage–host interactions. Together, these frameworks have given rise to influential ecological theories, such as Kill-the-Winner and Piggyback-the-Winner, which offer mechanistic explanations for phage–host dynamics. Despite these advances, theoretical insights have remained only loosely connected to environmental observations, and studies of phage communities have often been limited to descriptive characterization. Metagenomics now provides a unique opportunity to directly observe phage–host dynamics across diverse environments and timescales. The increasing availability of large-scale and longitudinal metagenomic datasets enables the application of statistical and data-driven approaches, including co-occurrence and correlation analyses, network inference, dynamic modeling, and machine learning, to infer interactions and ecological strategies from empirical data. In this review, we synthesize the foundations of theoretical and mathematical modeling of phage–host systems and discuss emerging approaches for integrating omics data with these frameworks. By linking metagenomic observations with mechanistic and ecological models, we highlight pathways toward moving beyond descriptive viromics and toward predictive, data-informed understanding of phage ecology and evolution in natural ecosystems.