Integration of marine meta-omics data through genome-scale metabolic models

Marine microbial communities play a central role in biogeochemical cycling, yet linking genomic potential and transcriptional activity to quantitative metabolic function remains a major challenge, particularly for uncultivated taxa. Genome-scale metabolic models (GEMs) provide a mechanistic framework to bridge this gap, but their application to environmental microorganisms requires careful integration of multi-omic data. In this thesis, an integrative modeling approach is developed and applied to investigate the metabolic behavior of Lentibacter algarum, a marine heterotrophic bacterium within the Roseobacteraceae family associated with algal blooms.​ ​ Using a time series sampled during a phytoplankton bloom in Monterey Bay, genome-resolved metagenomics was employed to reconstruct high quality metagenome-assembled genomes (MAGs), from which the first metabolic model of L. algarum was generated and curated. Metatranscriptomic data were then integrated into the model to produce condition-specific metabolic representations constrained with expression data, adapting existing transcriptomics integration strategies to the characteristics of marine environmental data. Flux balance analysis was applied across multiple samples to explore transcriptionally driven changes in metabolic flux distributions.​ ​ The resulting L. algarum model revealed distinct metabolic states associated with biomass growth and environmental context. Transcriptomics informed models displaying high levels of biomass growth were characterized by an increased flux through biosynthetic pathways, particularly within carbon and energy metabolism, whereas models displaying low levels of biomass growth were associated with a broader activation of metabolic subsystems consistent with maintenance-oriented strategies. Furthermore, chlorophyll concentration, used as a proxy for the bloom dynamics, was associated with a systematic reconfiguration of metabolite fluxes, highlighting the functional responses of L. algarum to phytoplankton-derived resource availability.​ ​ Overall, this work demonstrates that integrating metagenomic and metatranscriptomic data within a genome-scale metabolic modeling framework provides enhanced insight into the metabolism of marine bacteria in dynamic environments. By linking transcriptional regulation to metabolic behavior at the systems level, this thesis contributes a quantitative, process-oriented perspective on microbial functioning during phytoplankton blooms and outlines a broadly applicable framework for transcriptomics informed metabolic modeling in microbial ecology.