Genomic Science Program
U.S. Department of Energy | Office of Science | Biological and Environmental Research Program

Microbes Persist: Towards Quantitative Theory-Based Predictions of Soil Microbial Fitness, Interaction, and Function in Knowledgebase


Gianna Marschmann1* (, Ulas Karaoz1, Jeffrey Kimbel2, Ben Koch3, Steven J. Blazewicz2, Bruce Hungate3, Eoin Brodie1, Jennifer Pett-Ridge2,4


1Earth and Environmental Sciences, Lawrence Berkeley National Laboratory; 2Physical and Life Sciences Directorate, Lawrence Livermore National Laboratory; 3Center for Ecosystem Science and Society, Northern Arizona University–Flagstaff; 4University of California–Merced



Microorganisms play key roles in soil carbon (C) turnover and stabilization of persistent organic matter via their metabolic activities, cellular biochemistry, and extracellular products. Microbial residues are the primary ingredients in soil organic matter (SOM), a pool critical to Earth’s soil health and climate. The team hypothesizes that microbial cellular-chemistry, functional potential, and ecophysiology fundamentally shape soil C persistence, and the researchers are characterizing this via stable isotope probing (SIP) of genome-resolved metagenomes and viromes. This study focuses on soil moisture as a “master controller’” of microbial activity and mortality since altered precipitation regimes are predicted across the temperate United States. The Science Focus Area’s (SFA’s) ultimate goal is to determine how microbial soil ecophysiology, population dynamics, and microbe-mineral-organic matter interactions regulate the persistence of microbial residues under changing moisture regimes


Researchers have developed a genomes-to-traits workflow (microTrait) and a compatible dynamic energy budget trait-based model (DEBmicroTrait) to (1) infer ecologically relevant traits from microbial genomes; (2) systematically reduce the high-dimensionality of genome-level microbial trait data by inferring functional guilds (sets of organisms performing the same ecological function irrespective of their phylogenetic origin); (3) quantify within-guild trait variance and capture trait linkages in trait-based models; and (4) explore trait-based simulations under different scenarios with varying levels of microbial community and environmental complexity (Karaoz and Brodie 2022; Marschmann et al. 2024). This computational workflow allows the team to predict trade-offs involving metabolic, biophysical, and thermodynamic traits of microorganisms. This includes the capability to predict substrate uptake kinetics for broad substrate classes. Researchers are working to integrate this tool within KBase, which will support ongoing efforts to integrate genome-centric knowledge into biogeochemical models.

In addition, the SFA team has formalized and optimized the code for quantitative stable isotope probing (qSIP) into an R package (qSIP2) and documentation website (Hungate et al. 2015; Koch et al. 2018; Simpson et al. 2023). The qSIP2 workflow (and forthcoming KBase applications) allow for the identification of enriched taxa in isotope addition experiments given density fractionation of DNA, sequence counts, and a quantitative measure of abundance in each fraction (e.g. 16S rRNA). The qSIP2 workflow can accept input for both amplicon (e.g. 16S rRNA) and metagenomic (e.g., MAG coverage) data and produce an excess atom fraction enrichment value quantifying the extent of “heavy” isotope incorporation. Results from qSIP can help experimentally identify microbial traits via quantifying the use of a given substrate.

Ongoing work to combine both the qSIP and (DEB)microTrait tools within KBase will provide a strong foundation for researchers who wish to use quantitative in situ measurements of microbial ecophysiology and population dynamics to benchmark models and build a predictive understanding of biological processes controlling material fluxes in complex environments.


Hungate, B.A., et al., 2015. “Quantitative Microbial Ecology Through Stable Isotope Probing,” Applied and Environmental Microbiology 81, 7570–81.

Karaoz, U., and E. L. Brodie. 2022. “MicroTrait: a Toolset for a Trait-Based Representation of Microbial Genomes,” Frontiers in Bioinformatics 918853.

Koch, B. J., et al. 2018. “Estimating Taxon-Specific Population Dynamics in Intact Microbial Communities,” Ecosphere 9, e02090.

Marschmann, G.L.., et al. 2024. “Predictions of Rhizosphere Microbiome Dynamics with a

Genome-Informed and Trait-Based Energy Budget Model,” Nature Microbiology. 1–13.

Simpson, A., et al. 2023. “A Data Standard for the Reuse and Reproducibility of Any Stable Isotope Probing-Derived Nucleic Acid Sequence (MISIP),” bioRxiv 2023(7).

Funding Information

This research is based upon work of the Lawrence Livermore National Laboratory (LLNL) “Microbes Persist” Soil Microbiome Science Focus Area, supported by the DOE Office of Science, BER program, Genomic Science program under Award Number SCW1746 to the Lawrence Livermore National Laboratory. Work was carried out at LLNL under DOE Contract DE-AC52-07NA27344, at Lawrence Berkeley National Laboratory under DOE Contract DE-AC02- 05CH11231, and at Northern Arizona University.