Bridging the gap between omics and earth system science to better understand how environmental change impacts marine microbes

Abstract

The advent of genomic-, transcriptomic- and proteomic-based approaches has revolutionized our ability to describe marine microbial communities, including biogeography, metabolic potential and diversity, mechanisms of adaptation, and phylogeny and evolutionary history. New interdisciplinary approaches are needed to move from this descriptive level to improved quantitative, process-level understanding of the roles of marine microbes in biogeochemical cycles and of the impact of environmental change on the marine microbial ecosystem. Linking studies at levels from the genome to the organism, to ecological strategies and organism and ecosystem response, requires new modelling approaches. Key to this will be a fundamental shift in modelling scale that represents micro-organisms from the level of their macromolecular components. This will enable contact with omics data sets and allow acclimation and adaptive response at the phenotype level (i.e. traits) to be simulated as a combination of fitness maximization and evolutionary constraints. This way forward will build on ecological approaches that identify key organism traits and systems biology approaches that integrate traditional physiological measurements with new insights from omics. It will rely on developing an improved understanding of ecophysiology to understand quantitatively environmental controls on microbial growth strategies. It will also incorporate results from experimental evolution studies in the representation of adaptation. The resulting ecosystem-level models can then evaluate our level of understanding of controls on ecosystem structure and function, highlight major gaps in understanding and help prioritize areas for future research programs. Ultimately, this grand synthesis should improve predictive capability of the ecosystem response to multiple environmental drivers.

Keywords: evolution; genomics; microbes; modelling; ocean.

Figure 1

Informing the Earth system science with marine phytoplankton by omics data. Metatranscriptome sequences from natural phytoplankton communities helped to identify physiological traits (cellular concentration of ribosomes and their rRNAs) underpinning adaptation to environmental conditions (temperature). A mechanistic phytoplankton cell model was used to test the significance of the identified physiological trait for cellular stoichiometry. Environmental selection in a trait‐based global marine ecosystem model was then linking emergent growth and cellular allocation strategies to large‐scale patterns in light, nutrients and temperature in the surface marine environment. Global predictions of cellular resource allocation and stoichiometry (N:P ratio) were consistent with patterns in metatranscriptome data (Toseland et al., 2013) and latitudinal patterns in the elemental ratios of marine plankton and organic matter (Martiny et al., 2013). Three‐dimensional view of ribosome was taken from Wikipedia, showing rRNA in dark blue and dark red. Lighter colours represent ribosomal proteins. Bands above show temperature‐dependent abundance of the eukaryotic ribosomal protein S14, adapted from Toseland et al. (2013).

Figure 2

Bridging the gap: a model‐centred approach to integrating omics approaches with marine microbial ecology. Omics approaches (blue bars) provide new insights both at the level of population and community structure (red bar), and into physiology at organism level (green bar) and below. Quantitatively understanding ecosystem structure, function and response to environmental change requires both integration of omics approaches with other methods and a hierarchical forward (or ‘bottom‐up’) modelling approach (blue arrows). This first links omics to physiology via a combination of gene‐scale models (metabolic networks, transcriptional regulation) and whole‐cell models that represent transport processes, storage pools and energetics. It then represents selection in a model environment to predict community composition and function from organism traits. Evaluation against the combination of omics and other data sets (including satellite colour, in situ nutrient and rate measurements) then indicates missing processes. Including a model representation of genetic constraints on adaptation (microevolution) derived from laboratory experimental evolution studies and observed genetic diversity and structure then enables a predictive model for response to environmental change.