Design and Omics Exploration of Synthetic Microbial Communities in KBase
Authors:
Priya Ranjan1* (ranjanp@ornl.gov), Andrew Freiburger2, Alexis L. Marsh3, Dale A. Pelletier1, Myra B. Cohen3, David J. Weston1, Christopher S. Henry2, Paramvir S. Dehal4, Adam Arkin4, Robert Cottingham1, Mitchel J. Doktycz1
Institutions:
1Oak Ridge National Laboratory; 2Argonne National Laboratory; 3Iowa State University; and 4Lawrence Berkeley National Laboratory
Goals
Simple constructed communities with desired biological functions can be used to study bacterial processes involved in community establishment and mimic the behavior of natural communities to increase plant growth and disease resistance. In this collaboration with the Plant-Microbe Interfaces (PMI) science focus area (SFA) at Oak Ridge National Laboratory, team members are adding datasets and apps to KBase to simplify the selection of isolates for constructed community experiments. For designing the community, researchers can use KBase Apps that annotate genomes with plant growth–promoting traits and secondary metabolism classes as well as an app that calculates metabolic dependencies between microbes of interest. The design process will be tested using experimental systems established in the PMI SFA for studying constructed communities. Further, the results from these experiments will be integrated back into KBase to improve the mechanistic understanding of interactions and iteratively improve the design process.
Abstract
Simple constructed communities offer a key experimental platform for examining how environmental perturbations affect the structure of the microbiome and host physiology and productivity. Computational simulations are a necessary tool to guide the design of simplified constructed communities. KBase—a DOE BER-funded, public, and freely accessible software and data science platform with a rich user interface—is ideal for developing such computational tools since it already offers a large and increasing number of diverse tools: e.g., functional annotation, metabolic modeling, auxotrophy prediction, substrate utilization and production of byproducts, taxonomic information, and predicting microbial traits. With these tools, it is possible to get some insights about a genome’s biochemistry and general characteristics only based on its sequence. The following user interface applications (apps) are under development to expand the set of KBase tools in support of constructed community studies.
The first app, Annotate Genomes with Plant Growth Promoting Traits, applies annotation to genes in a genome based on the Plant Growth Promoting Trait ontology (PGPT). PGPT is a literature- and omics-curated, comprehensive, and hierarchical collection containing 6,900 PGPTs associated with 6,965,955 protein sequences. The ontology has several categories such as phytohormone production, plant signal production, bio-fertilization (potassium solubilization, iron acquisition), bioremediation (fluoride, heavy metal detoxification), colonizing plant system (chemotaxis, surface attachment, root colonization), plant immune response stimulation, stress control, competitive exclusion (quorum sensing, bacterial fitness, cell envelope remodeling). This app can help a user prioritize and select genomes for constructed community experiments.
The app Annotate Genomes with Secondary Metabolism Classes uses antiSMASH (1), a popular genomics tool, to annotate genes in a genome with secondary metabolism classes and identify biosynthetic gene clusters within the microbial genomes of interest. The biosynthetic gene cluster profile can be used to generate hypotheses regarding certain organisms having a higher potential for antagonistic and antimicrobial activity. This information can then be used to form subsequent hypotheses regarding the membership of stable communities. The team will also be exploring outputs from the JGI Secondary Metabolite Collaboratory as they work with them in the integration of the analysis workflow into KBase.
The app Calculate Metabolic Interaction Score builds pairwise metabolic models in a given environment and calculates several quantitative metrics that describe competitive and cooperative potential between the paired microbes. These metrics include: (1) the metabolic interaction potential (MIP), which approximates the potential for syntrophic cooperation between the organisms; (2) the metabolic resource overlap (MRO), which approximates competition for media substrates between the organisms; and (3) predicted growth rates of the isolates and community, which leverages other KBase Apps to approximate growth dynamics of the given community.
The aforementioned apps will be used by PMI researchers to prioritize genomes for constructed community experiments and generate testable hypotheses: e.g., the apps can identify microbial functions and possibly redundant community members, which can be tested by replacing one member with another microorganism whose functional and metabolic profile are predicted to be similar or different. Team members have also uploaded 550 PMI isolate genomes to KBase and are exploring design of improved user interface for searching, filtering, and selecting genomes for constructed communities. These datasets and apps will expedite the design-build-test-learn cycles of PMI SFA projects and enable new scientific explorations.
References
Blin, K., et al. 2021. “antiSMASH 6.0: Improving Cluster Detection and Comparison Capabilities,” Nucleic Acids Research 49(1), W29–W35. DOI:10.1093/nar/gkab335.
Funding Information
Oak Ridge National Laboratory is managed by UT-Battelle, LLC for the U.S. Department of Energy under contract no. DE-AC05-00OR22725. This program is supported by the U. S. Department of Energy, Office of Science, through the Genomic Science Program, Office of Biological and Environmental Research, under FWP ERKPA39.