242 Chapter 8 the difficulty in modeling their growth using traditional Michaelis-Menten kinetics, as the environmental controls on electron flow are intricate and vary across systems. Functional marker gene modeling approaches typically require prior knowledge of associated markers from well-characterized metabolisms, which can limit their ability to capture novel and complex processes. The study by Louca et al. (2016) is notable for its validation of modeled anammox and denitrification processes. This validation was achieved by incorporating geochemical depth profiles and rate measurements, along with metagenomic, metatranscriptomic, and metaproteomic data, as well as quantitative PCR (qPCR) of specific microbial Gammaproteobacterial taxa. In this regard, one limitation of gene-centric model validation via meta-omics datasets is that they often provide relative rather than absolute abundances. For more precise cell quantification, techniques like bioorthogonal non-canonical amino acid tagging and fluorescence-activated cell sorting (BONCAT-FACS) could be employed to determine active cells per gram or aggregate. Active cell sorting methods also have limitations, including the need for specialized personnel with cell sorting expertise for in situ field preparation, biases in incubation conditions that may favor the selection of one taxon over others, and challenges related to complex sample preparation, such as autofluorescence, which can compromise the detection of active cell signals. In fluctuating ecosystems, metatranscriptomics is not always a reliable proxy for metabolic rates, as gene expression can vary over very short time scales (within minutes) (Moran et al., 2013). Instead, RNA-seq offers a highly sensitive indicator of ecologically relevant processes compared to metaproteomics (literature). Overall, to improve biogeochemical models in systems like the eutrophic Stockholm Archipelago in Chapter 2, incorporating multi-omics, qPCR-based functional gene abundance, and additional AOM inhibitory constants, will provide a more accurate representation of processes like S-AOM in such complex and dynamic environments. Additional S-AOM inhibitory sulfide incubations including contrasting sites such as euxinic (Site 3 Sandofjärden) and hypoxic (Lilla Vaärtan) environments, will further enhance the robustness of the models. In Chapter 2, including qPCR markers such as methyl coenzyme M reductase (mcrA) (specific
RkJQdWJsaXNoZXIy MTk4NDMw