Breaking the reproducibility barrier with standardized protocols for plant–microbiome research

Vlastimil Novak; Peter F Andeer; Eoghan King; Jacob Calabria; Connor Fitzpatrick; Jana M Kelm; Kathrin Wippel; Suzanne M Kosina; Benjamin P Bowen; Chris Daum; Matthew Zane; Archana Yadav; Mingfei Chen; Dor Russ; Catharine A Adams; Trenton K Owens; Bradie Lee; Yezhang Ding; Zineb Sordo; Romy Chakraborty; Simon Roux; Adam M Deutschbauer; Daniela Ushizima; Karsten Zengler; Borjana Arsova; Jeffery L Dangl; Paul Schulze-Lefert; Michelle Watt; John P Vogel; Trent R Northen

doi:10.1371/journal.pbio.3003358

. 2025 Sep 8;23(9):e3003358. doi: 10.1371/journal.pbio.3003358

Breaking the reproducibility barrier with standardized protocols for plant–microbiome research

Vlastimil Novak ¹, Peter F Andeer ¹, Eoghan King ^2,³, Jacob Calabria ⁴, Connor Fitzpatrick ⁵, Jana M Kelm ⁶, Kathrin Wippel ^7,⁸, Suzanne M Kosina ¹, Benjamin P Bowen ^1,², Chris Daum ², Matthew Zane ², Archana Yadav ⁹, Mingfei Chen ⁹, Dor Russ ⁵, Catharine A Adams ^1,¹⁰, Trenton K Owens ¹, Bradie Lee ^2,¹⁰, Yezhang Ding ¹, Zineb Sordo ¹¹, Romy Chakraborty ⁹, Simon Roux ², Adam M Deutschbauer ^1,¹⁰, Daniela Ushizima ¹¹, Karsten Zengler ^12,^13,^14,¹⁵, Borjana Arsova ⁶, Jeffery L Dangl ⁵, Paul Schulze-Lefert ⁷, Michelle Watt ⁴, John P Vogel ^2,^*, Trent R Northen ^1,^2,^*

Editor: Cara Helene Haney¹⁶

¹Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, California, United States of America

²The DOE Joint Genome Institute, Lawrence Berkeley National Laboratory, Berkeley, California, United States of America

³Centro de Biotecnología y Genómica de Plantas, Universidad Politécnica de Madrid (UPM)–Instituto Nacional de Investigación y Tecnología Agraria y Alimentación (INIA/CSIC), Campus de Montegancedo, Madrid, Spain

⁴Faculty of Science, School of BioSciences, The University of Melbourne, Parkville, Australia

⁵Department of Biology and Howard Hughes Medical Institute, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America

⁶Institute for Bio- and Geosciences, Plant Sciences (IBG-2), Forschungszentrum Jülich GmbH, Jülich, Germany

⁷Max Planck Institute for Plant Breeding Research, Cologne, Germany

⁸Swammerdam Institute for Life Sciences, University of Amsterdam, Science Park Amsterdam, Amsterdam, The Netherlands

⁹Earth and Environmental Sciences, Lawrence Berkeley National Laboratory, Berkeley, California, United States of America

¹⁰Department of Plant and Microbial Biology, University of California-Berkeley, Berkeley, California, United States of America

¹¹Computing Sciences Area, Lawrence Berkeley National Laboratory, Berkeley, California, United States of America

¹²Department of Pediatrics, University of California, San Diego, La Jolla, California, United States of America

¹³Department of Bioengineering, University of California, San Diego, La Jolla, California, United States of America

¹⁴Center for Microbiome Innovation, University of California, San Diego, La Jolla, California, United States of America

¹⁵Program in Materials Science and Engineering, University of California, San Diego, La Jolla, California, United States of America

¹⁶University of Pittsburgh Dietrich School of Arts and Sciences, UNITED STATES OF AMERICA

^✉

* E-mail: jpvogel@lbl.gov (JPV); trnorthen@lbl.gov (TRN)

I have read the journal’s policy and the authors of this manuscript have the following competing interests: P.F.A. and T.R.N. are inventors of patent US11510376B2, held by the University of California, covering an Ecosystem device for determining plant-microbe interactions. All other authors have declared that no competing interests exist.

Roles

Vlastimil Novak: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

Peter F Andeer: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing

Eoghan King: Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing

Jacob Calabria: Data curation, Investigation, Writing – review & editing

Connor Fitzpatrick: Data curation, Investigation, Writing – review & editing

Jana M Kelm: Data curation, Investigation, Writing – review & editing

Kathrin Wippel: Data curation, Investigation, Writing – review & editing

Suzanne M Kosina: Formal analysis, Validation, Writing – review & editing

Benjamin P Bowen: Formal analysis, Writing – review & editing

Chris Daum: Investigation, Resources, Writing – review & editing

Matthew Zane: Investigation, Resources, Writing – review & editing

Archana Yadav: Formal analysis, Visualization, Writing – review & editing

Mingfei Chen: Investigation, Writing – review & editing

Dor Russ: Investigation, Writing – review & editing

Catharine A Adams: Resources, Writing – review & editing

Trenton K Owens: Investigation, Writing – review & editing

Bradie Lee: Methodology, Writing – review & editing

Yezhang Ding: Investigation, Writing – review & editing

Zineb Sordo: Data curation, Formal analysis, Writing – review & editing

Romy Chakraborty: Funding acquisition, Resources, Supervision, Writing – review & editing

Simon Roux: Data curation, Formal analysis, Supervision, Writing – review & editing

Adam M Deutschbauer: Formal analysis, Funding acquisition, Supervision, Writing – review & editing

Daniela Ushizima: Data curation, Formal analysis, Supervision, Writing – review & editing

Karsten Zengler: Conceptualization, Funding acquisition, Supervision, Writing – review & editing

Borjana Arsova: Data curation, Supervision, Writing – review & editing

Jeffery L Dangl: Data curation, Supervision, Writing – review & editing

Paul Schulze-Lefert: Data curation, Supervision, Writing – review & editing

Michelle Watt: Data curation, Resources, Supervision, Writing – review & editing

John P Vogel: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Trent R Northen: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Cara Helene Haney: Academic Editor

PMCID: PMC12416739 PMID: 40920748

Abstract

Inter-laboratory replicability is crucial yet challenging in microbiome research. Leveraging microbiomes to promote soil health and plant growth requires understanding underlying molecular mechanisms using reproducible experimental systems. In a global collaborative effort involving five laboratories, we aimed to help advance reproducibility in microbiome studies by testing our ability to replicate synthetic community assembly experiments. Our study compared fabricated ecosystems constructed using two different synthetic bacterial communities, the model grass Brachypodium distachyon, and sterile EcoFAB 2.0 devices. All participating laboratories observed consistent inoculum-dependent changes in plant phenotype, root exudate composition, and final bacterial community structure, where Paraburkholderia sp. OAS925 could dramatically shift microbiome composition. Comparative genomics and exudate utilization linked the pH-dependent colonization ability of Paraburkholderia, which was further confirmed with motility assays. The study provides detailed protocols, benchmarking datasets, and best practices to help advance replicable science and inform future multi-laboratory reproducibility studies.

Reproducing microbiome experiments across laboratories is a major challenge in soil and plant health research. In a collaborative effort involving five laboratories, this study achieved consistent, inoculum-dependent results, providing detailed protocols, datasets and best practices to promote reproducibility in plant microbiome studies.

1. Introduction

As recent perspective papers have highlighted, establishing model microbiomes is a pressing need in environmental microbiology [1,2]. Several years ago, a vision was presented for developing and validating standardized ‘fabricated ecosystems’ to enable replicable studies of microbiomes in ecologically relevant contexts, akin to the adoption of shared model organisms [1]. A fabricated ecosystem is defined as a closed laboratory ecological system where all biotic and abiotic factors are initially specified/controlled. Synthetic microbial communities (SynComs) are valuable tools for bridging the gap between natural communities and studies involving axenic cultures and isolates [3]. By limiting complexity yet retaining functional diversity and microbe-microbe interactions, SynComs can be used to unravel mechanisms underlying complex interactions, providing critical insights into community assembly processes, microbial interactions, and host physiology, e.g., plant host [3–6]. These interactions between the host and its microbes define the holobiont concept, where the plant and its microbiome form a single dynamic ecological unit [7]. However, standardization is essential to fully leverage the potential of SynComs and achieve replicable plant microbiome studies [8]. This requires overcoming several challenges, including the availability of strains and standardized protocols for their growth in the laboratory. To address these challenges, we recently developed a standardized model community of 17 bacterial isolates from a grass rhizosphere available through a public biobank (DSMZ), along with cryopreservation and resuscitation protocols [9].

Other aspects to enable replicable microbiome studies must be standardized, including sterile habitats and protocols for sample collection and analysis [1]. As initial steps towards this vision, we developed a first-generation sterile container for fabricated ecosystems (EcoFAB device) and performed a multi-laboratory study demonstrating the reproducible physiology of the model grass Brachypodium distachyon [10]. Recently, it was found that Paraburkholderia sp. OAS925 dominated other members of the model 17-member SynCom for B. distachyon root colonization [11]. Additionally, we have since developed an improved EcoFAB 2.0 device that enables highly reproducible plant growth [12]. The next step towards standardization is to test the replicability of microbiome formation, plant responses to microbiomes, and root exudation using these standardized laboratory habitats and SynComs. This can be achieved through inter-laboratory comparison studies or ring trials—a powerful tool in proficiency testing of analytical methods [13,14] that are currently underutilized in microbiome research.

Here, we describe a five-laboratory international ring trial investigating the reproducibility of B. distachyon phenotypes, exometabolite profiles, and microbiome assembly within the EcoFAB 2.0 device. The experiment compared the recruitment of the full SynCom versus one lacking the dominant root colonizer Paraburkholderia sp. OAS925 [11]. To minimize variation required in all laboratories, almost all supplies (e.g., EcoFABs 2.0, seeds, SynCom inoculum, filters) were distributed from the organizing laboratory, and detailed protocols, including annotated videos, were created. Each laboratory measured plant phenotypes and collected samples for 16S rRNA amplicon sequencing and metabolomic analyses by LC–MS/MS. A single laboratory performed all the sequencing and metabolomic analyses to minimize analytical variation. Follow-up in vitro assays and comparative genomics were conducted to gain insights into mechanisms leading to Paraburkholderia sp. OAS925 dominance. Overall, the study demonstrates consistent plant traits across multiple laboratories and provides publicly accessible benchmarking data for other labs to leverage, replicate, and extend this work. In addition, we describe the challenges we encountered in performing this study, thus providing information that can facilitate future microbiome reproducibility studies.

2. Results

2.1. Study design and logistics

Our main objective was to develop and test methods to reproducibly study plant microbiomes within the sterile EcoFAB 2.0 device (Fig 1a). We hypothesized that the inclusion of Paraburkholderia sp. OAS925, a dominant B. distachyon root colonizer into SynCom [11], would reproducibly influence the microbiome assembly, metabolite production, and plant growth across multiple laboratories using the EcoFAB 2.0 device. To test the hypothesis, we deployed the grass B. distachyon with a SynCom consisting of 16 or 17 members (either with or without Paraburkholderia sp. OAS925) that was originally developed to span the diversity of bacteria isolated from a grass rhizosphere, including representatives from the Actinomycetota, Bacillota, Pseudomonadota, and Bacteroidota phyla (Fig 1b) [9]. Our study was conducted across five laboratories (designated A–E) and consisted of four treatments with seven biological replicates each (Fig 1a): an axenic (mock-inoculated) sterile plant control, SynCom16-inoculated plants, SynCom17-inoculated plants, and plant-free medium control. Each laboratory followed written protocols and annotated videos, gathered root and unfiltered media samples for 16S rRNA amplicon sequencing, filtered media for metabolomics, measured plant biomass, and performed root scans. At the end of the study, the collected data and samples were sent to the organizing laboratory for sequencing, metabolomics, and data analysis.

The detailed protocol with embedded annotated videos used by all five laboratories is available via protocols.io (https://dx.doi.org/10.17504/protocols.io.kxygxyydkl8j/v1) [15]. The general procedure follows these steps: (i) EcoFAB 2.0 device assembly; (ii) B. distachyon seed dehusking, surface sterilization, and stratification at 4 °C for 3 days; (iii) Germination on agar plates for 3 days; (iv) Transfer of seedlings to the EcoFAB 2.0 device for an additional 4 days of growth; (v) Sterility test and SynCom inoculation into the EcoFAB 2.0 device; (vi) Water refill and root imaging at three timepoints; (vii) Sampling and plant harvest at 22 days after inoculation (DAI). Since differences in labware and material can cause experimental variation, the protocol specifies the part numbers used in this study. Organizers provided critical components, including growth chamber data loggers, in the initial package of nonperishable supplies, while the SynComs and freshly collected seeds were shipped just before the study. Given the time zone differences, it was difficult to synchronize all activities, so each laboratory performed the experiment independently within 1.5 months of each other (S1 Table). All participants followed data collection templates and image examples.

2.2. Protocols resulted in reproducibly sterile conditions

During the study, all participating laboratories tested the sterility of the EcoFABs 2.0 devices by incubating spent medium on Luria–Bertani (LB) agar plates at two different time points. Less than 1% (2 out of 210) of all tests showed colony formation (Fig 2a). Namely, a single colony was observed in one treatment of laboratory D in SynCom17, and multiple colonies for laboratory B in medium-only control (plate had cracked lid).

Fig 2 — (a) Sterility of uninoculated EcoFAB 2.0 devices tested across laboratories A–E and treatments (Trt.: Ax-Axenic, 16-SynCom16, 17-SynCom17, Tx-Medium technical control) at day 0 and day 22 after inoculation (DAI). The medium from these devices was incubated on LB agar plates for 22 days to observe bacterial colony formation. Each square represents one test with gray fill indicating sterility and black contamination. The photos of the agar sterility test can be found at https://doi.org/10.6084/m9.figshare.26409220. (b) Plant biomass weight combined across all laboratories (Lab A–E in different colors), measured as shoot dry weight, shoot and root fresh weight. One-way ANOVA with Tukey test, n = 7, ns p > 0.5, *p < 0.05, **p < 0.01, ***p < 0.001. The shoot photos can be found at https://doi.org/10.6084/m9.figshare.26409310. (c) Root system development was analyzed using RhizoNet (Lab B–E) and ImageJ (Lab A). The raw root pixel counts were normalized to the maximum value in each lab. Two-way ANOVA with Dunnet’s test vs. Axenic control, n = 7, ns p > 0.5, *p < 0.05, **p < 0.01, ***p < 0.001. The root scans and Rhizonet reports can be found at https://doi.org/10.6084/m9.figshare.26131291. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

2.3. Reproducible plant growth

When plant biomass data were combined across laboratories, we observed a significant decrease in shoot fresh weight and dry weight of plants inoculated with SynCom17, and to a lesser extent SynCom16, relative to the axenic treatment (Fig 2b). This said, we did observe some variability between laboratories (S1 Fig), which is presumably due to growth chamber differences including light quality (fluorescent versus LED growth lights), light intensity and temperature (S1 Table). Supporting this, the data loggers revealed variability in measured temperatures (S2a Fig) and photoperiod (S2b Fig). Image analysis of scanned roots (S3 Fig) revealed that SynCom17 caused a consistent decrease in root development observed after 14 DAI onwards (Fig 2c).

2.4. Reproducible microbiome assembly

SynComs were prepared using optical density at 600 nm (OD₆₀₀) to colony-forming unit (CFU) conversions (S2 Table) to ensure equal cell numbers (final inoculum 1 × 10⁵ bacterial cells per plant) and shipped on dry ice to each laboratory as 100× concentrated stocks in 20% glycerol. The cells were resuspended and added to 10-day-old B. distachyon seedlings in the EcoFAB 2.0. After 22 days of growth, the roots and media were sampled, shipped back to the organizing laboratory, sequenced (see S3 Table for read counts), and compared to the original inoculum. For both SynComs (SynCom16 and SynCom17), the community composition at 22 DAI differed from the inoculum (Fig 3). As hypothesized, the root microbiome inoculated with SynCom17 was dominated by Paraburkholderia sp. OAS925 across all laboratories (98 ± 0.03% average relative abundance ± SD). In its absence (SynCom16), other isolates showed high relative abundance in the root microbiome with increased variability across laboratories, namely Rhodococcus sp. OAS809 (68 ± 33%), Mycobacterium sp. OAE908 (14 ± 27%), and Methylobacterium sp. OAE515 (15 ± 20%). The most dominant microbial isolates detected in root samples were also typically present in the media samples (S4a Fig). Ordination plots showed clear separations between SynCom16 and SynCom17 microbiomes for both root and media, with generally higher variability between samples for the SynCom16 microbiome (S4b Fig). There was a minimal contribution of unknown reads in all samples, consistent with the observed sterility of the controls. The three samples with the highest proportion of unknown reads (>2.5%) contained operational taxonomic units (OTUs) associated with human microbes (S4 Table), suggesting introduction during handling. Furthermore, SynCom17 treatment in laboratory D did not show unknown reads (S3 Table), suggesting that the failed sterility test (Fig 2a) was likely caused by plate downstream contamination.

Fig 3 — Microbiome composition of *Brachypodium distachyon* roots and starting inoculum of plants inoculated with SynCom16 or SynCom17 (±*Paraburkholderia* sp. OAS925). Letters indicate different laboratories, with each biological replicate shown (n = 7). The inoculum shows technical replicates (n = 3). The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315 or in S3 Table.

2.5. Reproducible rhizosphere metabolome

The spent medium from each fabricated ecosystem was filtered and shipped to the organizing laboratory for LC–MS/MS analysis (S5 Table), followed by targeted and untargeted metabolomics to determine the root exudate composition and metabolite profiles in the presence of different SynComs in the rhizosphere. The targeted analysis identified 60 metabolites spanning diverse metabolite classes (S6 Table). Hierarchical clustering revealed general clustering by treatment and not laboratory (Fig 4), consistent with the experimental reproducibility observed with plant growth phenotypes and root microbiome composition. Furthermore, the metabolite clustering showed several treatment-dependent metabolite changes. The first large cluster included diverse metabolites increased in the SynCom17 treatment. A second large cluster consisted of metabolites with lower relative concentrations in the SynCom17 treatment, represented mainly by amino acids. A third, much smaller cluster consisting primarily of nucleosides(tides) increased in the SynCom16 or both SynCom treatments. This finding highlights the prominent impact of the community dominated by Paraburkholderia sp. OAS925 on modulation of metabolite composition in the rhizosphere. This was further supported by untargeted metabolomics on 833 detected features that showed a clear separation between rhizosphere metabolomes of axenic plants and SynCom17, which was reproducible across all laboratories (S5 Fig). These changes may be due to metabolite production or uptake by the microbes or plant roots or the activity of extracellular enzymes [17–19].

Fig 4 — We show mean values for 60 identified metabolites for each lab/treatment combination (n = 7), row-normalized to the average sum peak height per lab. Row colors indicate metabolite classes. Cluster 1: Abundant in SynCom17; Cluster 2: Low in SynCom17; Cluster 3: Abundant in SynCom16 or both SynComs. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315 or in S6 Table.

2.6. Statistical evaluation of variation between labs

Ordination analysis using non-metric multidimensional scaling (NMDS) for microbiome (S4b Fig) and metabolome data (S5 Fig) showed a clear separation between treatments, but some separation between labs. To explore this in more detail, we quantified the lab- versus treatment-associated variation by analysis of variation (ANOVA) (S7 Table), pairwise tests (S8 Table), and coefficient of variation (CV).

ANOVA analysis of microbial abundance showed that more microbes were significantly different due to the treatments versus lab doing the experiment, across both roots and media (S6a Fig). When investigating the source of variation, we observed microbe-dependent results, with the lab accounting for <18% of the variation, while SynCom treatment had a larger contribution in several microbes, especially Praburkholderia, Rhodococcus, and Methylobacterium (S6b Fig). CV analysis showed consistent trends for both root and media samples, with SynCom16 displaying higher within-lab variability than SynCom17 in all labs (S6c Fig). Between-lab CVs were similar across treatments. Overall, the observed microbial abundance levels were more dependent on treatment than lab.

Metabolite levels were more affected by the lab-associated factors than the microbial abundances. However, treatment was still the primary influence in the majority of the metabolites. Both lab and treatment significantly affected the majority of the 60 identified metabolites (S7a Fig). Comparing these results between the various labs, we found that 7–23 metabolites were significantly different in specific pairwise lab comparisons, with lab pair A–B being the most different, and labs A–D, C–D, and B–D most similar to one another (S7b Fig). This said, most metabolites were primarily influenced by treatment rather than lab, though a subset, especially salsolinol, tyramine, and dopamine showed notable lab effects (S7c Fig). Metabolites in Axenic and SynCom17 treatments had lower CV than SynCom16 in all labs, and the within-lab CV was approximately the same as the between-lab CV (S7d Fig). Together, these results support the conclusion that metabolite abundances were largely treatment-driven, though some metabolites were sensitive to lab-specific differences.

2.7. Colonization by Paraburkholderia

Given the reproducible impacts of Paraburkholderia sp. OAS925 on our fabricated ecosystems, including plant growth phenotypes, microbiome structure, and rhizosphere exometabolites, we performed additional analyses to gain insights into potential mechanisms that may explain its dominance. Comparative genomic analysis showed that the Paraburkholderia sp. OAS925 genome (IMG/M Taxon ID: 2931840637) uniquely includes acid resistance genes such as glutamate and arginine transporters and decarboxylases (S8 Fig) and a gene module coding for a Type 3 Secretion System (T3SS), which was not found in any other member of the SynCom (S9 Table).

We inoculated B. distachyon with a red fluorescent protein (RFP) expressing Paraburkholderia sp. OAS925 to investigate spatial-temporal root colonization in EcoFAB 2.0. Clear RFP signals were detected at the root tip and in the maturation zone at 1 DAI, with increased biofilm formation observed at 3 DAI (S9 Fig). We noted both sessile colonies on the rhizoplane (S1 Video) and active swimming surrounding root cells (S2 Video).

The biofilm formation and motility observed during microscopy motivated the follow-up in vitro assays to further assess these characteristics across isolates. Paraburkholderia sp. OAS925 exhibited the sixth-highest biofilm formation and the most liquid-culture growth (S10 Fig), highlighting its potential to outgrow many other bacteria in SynCom17 [20].

Swimming motility assays on soft agar revealed that Paraburkholderia sp. OAS925 had the highest motility within the first 24 h (S11a Fig). Additionally, compared to other isolates with similar motility phenotypes (S11b Fig), it maintained fast swimming in acidic conditions (S11c Fig) in the range of the hydroponic medium (pH 5.5–6.0 at the start of the experiment). KEGG mapping (map02040) showed the presence of flagellar assembly genes, suggesting that the observed motility is due to flagella.

3. Discussion

There is an urgent need to move towards replicable experimental systems to address common difficulties in reproducing microbiome experiments [2,21]. Here, we report what, to our knowledge, is the first multi-laboratory microbiome reproducibility study. We constructed fabricated ecosystems using two SynComs, the model plant B. distachyon Bd21-3, and sterile EcoFAB 2.0 devices. These, in combination with written protocols and annotated videos, generally resulted in reproducible plant growth phenotypes, host microbiomes, and exometabolomes across five laboratories spanning three continents. Of the 210 sterility tests performed across the 5 labs, only 2 potential contamination events were detected (Fig 2a). One of these was from a plate with a cracked lid, and so we conclude that the EcoFAB 2.0 devices and protocols were effectively able to achieve sterility across all labs.

The magnitude of laboratory-specific effects on microbiome and metabolome data varied across our study, with laboratory factors explaining <18% of the variation in 16S rRNA sequencing data, but having a greater impact on specific exometabolites, where 15 metabolites had variation >18% (S7 Table). For example, hexosaminic acid, deoxycytidine, and N-acetylputrescine were highly reproducible between labs, whereas salsolinol, tyramine, and dopamine showed greater variability. The biochemical connection between these metabolites (salsolinol derived from dopamine, which is synthesized from tyramine [22]) and dopamine’s instability (prone to oxidation [23]) may contribute to their high variability across laboratories. These findings suggest opportunities to reduce metabolite variability through refined sample collection, handling, and storage. Together, these findings emphasize the importance of accounting for laboratory-specific effects when interpreting metabolome data, and highlight the need for standardized methods to minimize variability to generate high-quality data that can be integrated across labs (e.g., using AI).

Our study confirms previous findings that Paraburkholderia sp. OAS925 dominates the root microbiome of B. distachyon when part of the SynCom [11]. Similar results were observed for another grass, Avena barbata, grown in its native soil, where members of the order Burkholderiales were the most active bacteria in the rhizosphere based on carbohydrate depolymerization [24]. Soil pH, organic carbon availability, oxygen levels and redox status are key factors influencing microbial community composition [25]. Our study suggests several traits that likely help contribute to the dominance of Paraburkholderia sp. OAS925 in SynCom17. For example, its high motility in acidic environments (S11c Fig), such as the rhizosphere, might facilitate quick colonization of ecological niches and affect community assembly [26–28]. Its ability to utilize amino acids like arginine, glutamine, and glutamate (Fig 4) provides a possible mechanism for cytoplasm de-acidification to maintain motility-enabling transmembrane proton gradient [29,30]. These results align with a previous study showing that Pseudomonas simiae genes involved in motility, carbohydrate metabolism, cell wall biosynthesis, and amino acid transport aid in colonization of Arabidopsis roots [31]. Interestingly, in SynCom16, Rhodococcus often dominates on roots (Fig 3) and shares fast growth (S10 Fig) and high motility (S11a Fig) with Paraburkholderia.

Both SynCom16 and SynCom17 decreased plant biomass, with SynCom17 having a more pronounced effect (Figs 2 and S1). The observed dominant colonization by Paraburkholderia (S9 Fig, S1 and S2 Videos) or other bacteria might disrupt plant nutrient homeostasis, as the root microbiome plays a crucial role in forming root diffusion barriers and maintaining plant mineral nutrient balance [5]. Furthermore, the observation of a T3SS (S9 Table) is consistent with previous findings in Paraburkholderia genomes and has been shown to play a role in root colonization and virulence. Future studies should investigate the role of T3SS in the dominance of Paraburkholderia sp. OAS925 in SynCom17 treatments and the associated plant biomass decrease. Additionally, future testing of the effects of Paraburkholderia sp. OAS925 and other dominant isolates on plant growth can provide additional insights into the extent that the observed growth effects are indeed a direct result of a single strain.

By organizing this ring trial, we learned valuable lessons that can be useful for future studies (Fig 5). First, it is important to perform pilot studies to optimize methods before initiating any multi-laboratory study. Long-distance sample and inoculum shipping posed challenges, especially given unpredictable long delays in customs and potential thawing due to dry ice sublimation [32]. Microorganism shipments require engagement with shippers and familiarity with country-specific import/export legal regulations. We also observed variability in plant biomass (Fig 2b), which could be attributed to differences, especially in light and temperature (S2 Fig) between the growth chambers used in each laboratory (S1 Table). Additionally, standardization of the scanner model for automated root system analysis could help reduce variability and improve reproducibility. Ideally, the same equipment would be used, with a real-time readout of environmental conditions, although this would significantly increase the cost of the study.

Despite our detailed protocols and annotated videos, several challenges remain to replicate microbiome studies, underscoring the importance of using the data from this study to benchmark future studies. We recommend using the comment section on protocols.io for ongoing refinement and clarification, allowing the procedure to evolve as a living document [33]. To provide FAIR (Findable, Accessible, Interoperable, and Reusable) data access and enable others to use these data for benchmarking, integration, and extension, all data from this study are available via the National Microbiome Data Collaborative (NMDC) project page (https://data.microbiomedata.org/details/study/nmdc:sty-11-ev70y104) [34]. The EcoFAB 2.0 devices (up to 50 pieces), B. distachyon Bd21-3 plant line, and metabolomics methods used in this study are currently freely available to domestic and international researchers via Joint Genome Institute (JGI) User Programs (https://jgi.doe.gov/), subject to a competitive scientific review process. The 16S rRNA sequencing is readily available via commercial and academic sequencing centers. Although the relative abundance of organisms should ideally not correlate with the sequencing facility, sample handling, DNA extraction, and bioinformatics can significantly impact results, underscoring the need to consider protocols when making comparisons [35]. Another challenge we see is that the strains of our SynCom are currently available as individual strains, so batch variation would be reduced if culture collections or private companies provided ready-to-use SynCom mixtures.

This study demonstrates that multiple geographically dispersed laboratories can reproduce SynCom-driven changes in plant phenotypes, community assembly, and exometabolite profiles. This was a challenging yet essential step in the vision outlined by Zengler and colleagues [1] to verify the reproducibility of experimental systems and protocols, which enable scientists to replicate and build on each other’s work. We see several ways these methods can help advance the field: first, scientists can replicate the study and compare their results against those reported here before extending the findings with additional modifications (e.g., adding phages, fungi, engineered strains, different hosts, new devices, etc.). Second, scientists can generate experimental data through replication and benchmarking, enabling integrative computational analyses that control laboratory-specific effects. Providing FAIR data and accompanying metadata and protocols, as done here, will be an essential step in achieving this vision. Such efforts would greatly enhance the application of machine learning to make generalizable discoveries drawn from multiple studies, ultimately leading to understanding microbial processes in complex natural environments.

4. Materials and methods

4.1. Preparation of synthetic bacterial communities

The bacterial isolates originate from the switchgrass rhizosphere and are available at DSMZ as individual strains [9] or as a collection (DSM 200000SY, inquire at contact@dsmz.de). Glycerol stocks had their 16S gene (27F - 1492R) sequenced to confirm isolate identity and purity. Each verified isolate was streaked on a preferred medium (S2 Table), and a single colony was inoculated into liquid culture. After 2–8 days at 27 °C, cultures were centrifuged at 5,000 g and washed with ½ Murashige and Skoog (MS) basal salts medium. The washed cultures were used to create SynComs by measuring OD₆₀₀, converting to CFU, and combining them to 2e7 cells/ml each (1:1 ratio) in 20% glycerol for cryopreservation [9]. We used SynComs composed of equal numbers of cells for the constituent strains, which is a widely used approach [9,36,37]. A 20% glycerol in ½ MS basal salts was a mock solution. Solutions were aliquoted (100 µL) in microcentrifuge tubes and stored at −80 °C. The participants diluted the solutions 100-fold, assuming 50% cell survival during freezing-thawing, resulting in a final theoretical CFU of 1e5/plant. Serial dilution established the OD₆₀₀ to CFU conversions, and a hemocytometer was used to measure Gottfriedia sp. OAE603.

The inoculums were distributed to five participants: Lawrence Berkeley National Laboratory (LBNL), USA (organizer); the University of Melbourne, Australia; the University of North Carolina at Chapel Hill, USA; Forschungszentrum Jülich, Germany; and the Max Planck Institute for Plant Breeding Research, Germany. Empty tubes were shipped first to estimate speed, dry ice sublimation, and select vendors. Shipments included content declaration, receiver cover letter, and country-specific permits. Identifying the closest phylogenetic species for each isolate helped avoid customs delays. Shipments to Germany were classified as biosafety level 1 (BSL1) and sent via FedEx International Priority with 8.2 kg of dry ice as a delivered-at-place with a proforma invoice detailing freight charges and goods value. The SynCom import to Australia required a permit for conditionally non-prohibited goods under the Biosecurity Act 2015 Section 179 (1) and was sent via Aeronet Worldwide with 22.68 kg of dry ice, which was refilled during the 9-day transit. The shipment to North Carolina used FedEx Standard Overnight with 9 kg of dry ice.

4.2. Experimental setup and plant growth conditions

Participating laboratories assembled EcoFAB 2.0 devices and followed the experimental protocol (https://dx.doi.org/10.17504/protocols.io.kxygxyydkl8j/v1) [15]. In summary, B. distachyon Bd21-3 seeds were surface-sterilized, plated on ½ MS basal salts with 1.5% phytoagar, and stratified for 3 days at 4 °C. The plates were then moved to the growth chamber with a 14 h photoperiod (16 h for lab C) at 26 °C and 10 h dark at 20 °C with photosynthetic photon flux density (PPFD) at 110–140 µmol/m²/s and 70% humidity if tunable. Each lab used two data loggers (HOBO MX2202 and UA-002-64) to track illuminance in lux normalized to maximum value per lab (for dark duration confirmation) and temperature. After 3 days, germinated seeds were transferred to sterile EcoFAB 2.0 devices with 9 mL of ½ MS basal salts (pH 5.5–6.0) and placed back in the chamber. After 4 days, plants were inoculated with Mock, SynCom16, or SynCom17, resuspended from 100 µL stocks to 10 ml using sterile ½ MS basal salts. Each EcoFAB 2.0 device received 1 ml of inoculum, resulting in a glycerol concentration of 0.02% in 10 ml of hydroponic medium. The treatment groups included: i) Mock-inoculated axenic plant control, ii) SynCom16-inoculated plants, iii) SynCom17-inoculated plants, and iv) Mock-inoculated technical control (plant-free). The sterility of the uninoculated devices was tested at 0 and 22 DAI by incubating hydroponic medium on LB agar plates for 22 days. Evaporated water was re-supplied and roots imaged by scanning every seven days. Some laboratories shifted their root imaging timepoints based on scanner availability. Plant harvest and sampling happened at 22 DAI.

4.3. Root phenotyping

To automate root analysis from flatbed scanner images, we used RhizoNet, a workflow for precise root segmentation, ideal for tangled roots in EcoFAB 2.0 devices [38]. Lab A’s scans faced issues with condensation and reflections, causing low contrast; thus, SmartRoot V4.21 was used instead [39]. The raw root measurements from both methods were normalized to the maximum value for each lab.

4.4. Samples collection and shipment

Sample collection and initial processing were performed independently in each lab. We collected 50 µL of the unfiltered growth medium for amplicon sequencing. The remaining spent medium was filtered via a 0.2 µm PES filter for metabolomic analysis, collected in polypropylene tubes (lab A: VWR 21008-103, labs B–E: VWR 93000-026) and stored at −80 °C before shipment. The root and shoot were separated during plant harvest, fresh weights were measured, roots were frozen for microbiome analysis by amplicon sequencing, and shoots were lyophilized for dry weight measurement. The filtered medium, unfiltered medium, and frozen roots were shipped to LBNL on dry ice with gel packs. The loggers were shipped separately at room temperature. To import intact frozen B. distachyon roots to the USA required a Controlled Import Permit (PPQ Form 588), supplier declaration, and TSCA Certification. Samples from Germany were shipped via DHL Medical Express (2 days) on 10 kg of dry ice, from Australia via Cryopdp (5 days) on 24 kg dry ice with a refill request, and from North Carolina via FedEx Priority Overnight on 9 kg dry ice (1 day).

4.5. Analysis of rhizosphere metabolites

At the LBNL, hydroponic medium samples were lyophilized (Labconco, Kansas City, MO). Empty tubes were used as extraction controls. During extraction, samples were kept on dry ice, and Solvents were chilled at −20 °C. The dried material was suspended in 1mL methanol (MX0486, Sigma), vortexed, and transferred to a microcentrifuge tube, followed by residue collection with an additional 0.5mL methanol. Samples were then sonicated (97043-944, VWR) in ice water for 15 min, centrifuged at 10,000g for 5 min at 10 °C, and supernatants transferred to new tubes. Supernatants were dried overnight by vacuum concentration (SpeedVac, Thermo). The following morning, samples were resuspended in 150 µL methanol containing internal standard mix (S5 Table), vortexed, and centrifuged at 10,000g for 5 min at 10 °C. Supernatants were filtered with 0.22 µm PVDF filters (UFC30GV, Millipore), transferred to amber glass vials with inserts (5,188−6,592, Agilent), and sealed with screw caps (5,185−5,820, Agilent). Samples were analyzed by LC-MS/MS, with polar metabolites separation using hydrophilic liquid interaction chromatography (column 683775-924, Agilent) on an Agilent 1290 HPLC system, followed by detection in positive ion mode on a Thermo Orbitrap Exploris 120 Mass Spectrometer (S5 Table). Samples were injected in positive mode, with methanol blanks between each sample; internal and external controls were used for quality control.

For untargeted metabolomics, the mzML files were processed via MZMine 3.0 [40] to create a feature list using a custom batch process (S1 File). The features with MS2 spectrum were then annotated in GNPS2 using spectral metabolite libraries [41]. Features with MS2, retention time (RT) > 0.6 min, and exudate sample max peak height >10× of extraction and technical controls were included, resulting in 833 features across all samples and labs. For targeted metabolomics, metabolites were identified (S6 Table) by analyzing the data with an in-house library of m/z, RT, and MS2 fragmentation information from authentic reference standards using Metabolite Atlas (https://github.com/biorack/metatlas) [42]. Only metabolites with a maximum exudate sample peak height >3× of extraction and technical controls were included. The identified metabolites were manually classified using the PubChem Classification Browser (https://pubchem.ncbi.nlm.nih.gov/classification/).

4.6. Microbiome analysis by amplicon sequencing

DNA was extracted from ground roots and media using the DNeasy PowerSoil Pro Kit (Qiagen), following the manufacturer’s instructions with minor modifications. Ground roots were mixed with a resuspension buffer, transferred to bead-beating tubes, and frozen at −80 °C. Samples were then thawed at 60 °C and bead-beaten using the FastPrep-24 system for 30 s at setting 5.0 (MP Biomedicals). The elution buffer pre-heated to 60 °C. Samples were extracted in batches with water-only samples as negative process controls. Glycerol stocks for 16- and 17-SynCom and a glycerol-only mock were included for time-zero data.

PCR amplification of V4 amplicons was performed in two steps. Library amplicons were generated using the Illumina i7 and i5 index/adapter sequences with V4 primers 515F (Parada) (GTGYCAGCMGCCGCGGTAA) and 806R (Apprill) (GGACTACNVGGGTWTCTAAT) [43]. Amplification used pooled primers on a Bio-RAD CFX 384 Real-Time PCR system with QuantiNova SYBR Green PCR kit (Qiagen) in 10 µL reactions with primers at 4 µM and mitochondrial and chloroplast PNA blockers (PNA Bio) at 1.25 µM. Amplification was initiated at 95 °C for 3 min, followed by the cycle: 95 °C for 8 s, 78 °C for 10 s, 54 °C for 5 s, and 60 °C for 30 s, followed by fluorescence measurement. Root samples were amplified for 22 cycles and media samples for 30 cycles with at least 1 replicate. To allow direct comparison, the SynCom16 and SynCom17 inocula were each amplified for both 22 and 30 PCR cycles, mirroring the amplification conditions used for root and media samples, respectively. Libraries were purified at least twice to remove excess primers using the Mag-Bind TotalPure NGS beads (0.8×), and targets were quantified using the QuantiFluor dsDNA System (Promega). Libraries were then pooled (i.e., root and media samples separately) and purified again. The concentration of the pooled library was determined using the Qubit dsDNA Assay Kit (BR or HS, Invitrogen). The flowcell was sequenced on the Illumina MiSeq sequencer using MiSeq Reagent kits, V3 (600-cycle), following a 2 × 00 indexed run recipe. MiSeq reads were processed using Usearch (v11.0.667) [44]. Reads were merged, trimmed, and short sequences removed using the ‘fastq_mergepairs, fastx_truncate, and fastq_filter’ commands. An initial OTU table was generated with the ‘fastx_uniques, cluster_otus, and otutab’ functions, and OTUs were assigned to SynCom members using the ‘annot’ function with V4 reference sequences. The three samples with the most unknown reads were analyzed via NCBI Microbial Nucleotide BLAST with Representative Genomes database and MegaBLAST algorithm to identify potential matches. The relative abundance of each SynCom member was calculated as the proportion of total microbial reads (excluding plant reads) per sample.

4.7. Growth and biofilm formation assays for bacterial isolates

Lab A conducted the biofilm formation assays. The crystal violet assay for biofilm formation was modified from a previous method [45]. Isolates were grown in R2A, washed, and resuspended in a 30 mM phosphate buffer. They were inoculated into the plates with NLDM medium [20] at a 1:10 (v/v) ratio (final volume of 100 µL and initial OD₆₀₀ of 0.02) and incubated statically at 30 °C for 3 days (n = 4–5) and growth was measured at OD₆₀₀. After incubation, wells were washed 3× with MilliQ water, air-dried, and stained with 125 µL of crystal violet solution (0.1% v/v crystal violet, 1% v/v methanol, and 1% v/v isopropanol in Milli-Q water) for 30 min at RT. Wells were rinsed 3× with MilliQ water, destained with 125 µL of 30% acetic acid for 30–60 min at RT, and absorbance at 550 nm (OD₅₅₀) of the destaining solution was measured.

4.8. Genomic analysis of bacterial isolates

All isolates were grown in R2A except Bradyrhizobium sp. OAE829, which was grown in 1/10 R2A. High molecular weight genomic DNA was extracted from bacterial pellets with the Monarch HMW DNA Extraction Kit (New England Biolabs) or the MasterPure Complete DNA Purification Kit (Lucigen). The genomic DNA was submitted to the JGI for sequencing using PacBio Sequel II or Illumina NovaSeq S4. Genomes were stored in the Genomes OnLine Database (GOLD) [46], followed by submission to the Integrated Microbial Genomes and Microbiomes (IMG/M) (https://img.jgi.doe.gov/) for annotation [47]. The annotated genomes can be accessed via Hugging Face (https://doi.org/10.57967/hf/5885) [16] or IMG/M under the listed Taxon ID or GOLD Project ID (S2 Table): Arthrobacter sp. OAP107 (2931867202, Gp0588953), Gottfriedia sp. OAE603 (2931797537, Gp0588949, formerly known as Bacillus sp. OAE603 [11]), Bosea sp. OAE506 (2931782253, Gp0589672), Bradyrhizobium sp. OAE829 (2931808876, Gp0589676), Brevibacillus sp. OAP136 (2931855177, Gp0588951), Paraburkholderia sp. OAS925 (2931840637, Gp0589681, formerly known as Burkholderia sp. OAS925 [9]), Chitinophaga sp. OAE865 (2931817136, Gp0589677), Lysobacter sp. OAE881 (2931823763, Gp0589678), Marmoricola sp. OAE513 (2931787146, Gp0589673), Methylobacterium sp. OAE515 (2931791092, Gp0589674), Mucilaginibacter sp. OAE612 (2931861231, Gp0588952), Mycobacterium sp. OAE908 (2852593896, Gp0440934), Niastella sp. OAS944 (2931847253, Gp0589682), Paenibacillus sp. OAE614 (2931801854, Gp0589675), Rhizobium sp. OAE497 (2931775946, Gp0589671), Rhodococcus sp. OAS809 (2931833612, Gp0589680), Variovorax sp. OAS795 (2931827682, Gp0588950).

Comparative genomics was conducted using genome statistics and KEGG pathways protein-coding gene abundance from IMG/M. Based on the genes involved in acid resistance of Escherichia coli, including Glutamine/Glutamate and Arginine membrane transporters (GadC and AdiC), glutaminase (YbaS), and decarboxylases (GadA, GadB, and AdiA) [29], we searched for genes annotated with similar functions (GltIJKL, HisPMQ-ArgT, GlnHPQ, glsA, GAD, AdiA). KEGG module coverage was compared between Paraburkholderia sp. OAS925 and the other 16 genomes using the “Statistical Analysis” tool from IMG [48] using Fisher’s exact test and modules with a corrected p-value <10⁻⁰⁵ were manually inspected for a potential link to plant root colonization.

A phylogenomic tree of the 17 SynCom members was constructed using the GTDB-tk workflow [49], incorporating 120 marker proteins for multiple sequence alignment. Fasttree [50] generated the tree, which included two closely related genomes for context, and it was visualized with Interactive Tree Of Life (iTOL) [51]. The phylum-level taxonomic classification is indicated for each member. Bootstrap values, derived from 100 replicates, are displayed for nodes with over 50 bootstrap support values. Chloroflexus aggregans served as an outgroup to root the tree.

4.9. Fluorescent microscopy

Lab A conducted the plant growth and microscopy. The pGinger plasmid 23100 containing the RFP gene under the kanamycin (Kan) resistance marker was introduced into Paraburkholderia sp. OAS925, as described previously [52]. Briefly, 1mL of Paraburkholderia sp. OAS925 grown overnight at 30 °C in R2A medium was mixed with 1 mL of E. coli S17 dapE-harboring the pGinger plasmid grown overnight at 37 °C on LB medium with Kan at 50 µg/mL and diaminopimelic acid (DAP) at 300 µM. The mixture was pelleted for 1 min at 10,000g and then resuspended in 100 µL water with 300 µM DAP. This mixture was then placed onto an R2A agar plate and incubated overnight at 30 °C. The bacterial mix was then scraped, resuspended in water, and plated on R2A with Kan 50 µg/mL. Transconjugants were verified via fluorescent microscopy and colony PCR.

The Paraburkholderia sp. OAS925 expressing RFP was grown overnight in 7 mL of R2A medium with Kan 20 mg/L, shaken at 200 rpm at 27 °C. The cells were pelleted by centrifugation at 4,000g for 10 min, resuspended in ½ MS basal salts, and used to inoculate 3-day-old plants in EcoFABs 2.0 at a starting OD₆₀₀ of 0.01. A flatbed scanner captured root architecture to indicate microscopy locations at 1 and 3 DAI. The EVOS M5000 imaging system (Thermo Fisher) was used for inverted microscopy, merging txRED and bright-field images. Uninoculated plants served as controls for autofluorescence.

4.10. Motility assays

Lab A conducted the motility assays. Pre-cultures were grown in 8 ml of liquid medium shaken at 200 rpm, at 27 °C in the dark, for 5–8 days. The swimming motility was tested by observing colony spreading on R2A soft agar plates (0.3% w/v). Initially, motility was tested for all 17 isolates at pH 7.2. Then, Paraburkholderia, Gottfriedia, or Brevibacillus were tested at pH 4, 5, 6, 7, 8, or 9 (adjusted with 1M HCl or 0.5M NaOH). Each 10 cm Petri dish containing 30 ml of solidified medium was inoculated with 5 µL of culture at the center (n = 3). Plates were incubated at 27 °C in the dark, and the motility ring diameter was measured after 24 and 45 h.

4.11. Data management, statistical analyses, software, and data visualization

The participating laboratories uploaded plant biomass data, root scans, and photos into a shared Google folder with structured directories and spreadsheets. The heat maps were generated with RStudio version 4.0.5 using heatmap.2 in the ggplots package [53]. The NMDS plots were also generated in RStudio using vegan package version 2.5-7 [54]. GraphPad Prism 10 version 10.2.3 generated all other plots and statistical analyses. Biorender.com was used to create graphical overviews. Microsoft Excel version 16.78.3 was used to store and manipulate data frames.

Statistical analysis of cross-laboratory variation included a two-way ANOVA by measuring significance of main effects and their interactions, calculation of source of variation as sum of squares of main effects and total sum of squares, and CV that was measured as a ratio between the standard deviation σ to the mean μ. The statistical analysis was applied to microbe abundance data from 16S rRNA sequencing and raw peak heights for 60 identified rhizosphere metabolites. Additionally, metabolite intensities were also analyzed by pairwise comparisons. The ANOVA and pairwise tests were done in Python v 3.8 using the Pingouin package version 0.5.5. The default ANOVA command was used between lab and treatment. The default pairwise tests command was used between lab and treatment with a p-value adjustment of Benjamini/Hochberg FDR correction. The code can be found at https://doi.org/10.6084/m9.figshare.29700029.

Supporting information

S1 Fig. Plant biomass data for laboratories A–E.

Box plots display all data points, with hinges spanning the 25th to 75th percentiles, a central line denoting the median, and whiskers reaching the minimum and maximum values. Different lowercase letters indicate statistically significant differences at p < 0.05. One-way ANOVA with Tukey test (n = 7). The shoot photos can be found at https://doi.org/10.6084/m9.figshare.26409310. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s001.tiff^{(924.7KB, tiff)}

S2 Fig. Plant growth conditions in labs A–E.

(a) Temperature (T) and (b) Normalized illuminance (% of lab maximum value) to assess lights-off duration, measured by HOBO loggers (Pendant model #UA-00264 in red, Bluetooth model #MX2202 in blue). Dashed lines show the set day/night T (26/20 °C); gray areas mark the 7-day pre-inoculation period. Labs A, B, and E experienced logging interruptions due to battery drainage; Lab D’s Bluetooth logger did not cover the experimental period. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s002.tiff^{(5.5MB, tiff)}

S3 Fig. Representative images of root systems.

Labs A–E imaged roots in EcoFAB 2.0 devices with flatbed scanners. The figure shows plants at harvest (22 DAI). The root scans and Rhizonet reports can be found at https://doi.org/10.6084/m9.figshare.26131291. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s003.tiff^{(8.6MB, tiff)}

S4 Fig. Microbiome composition.

(a) Microbiome in plant growth media and starting inoculum. Letters indicate different laboratories, with each biological replicate shown (n = 7). The inoculum shows technical replicates (n = 3). (b) NMDS plot of root and media microbiomes with 95% confidence ellipse. Different laboratories are shown with various symbols, while colors represent SynCom16 (blue) versus SynCom17 (orange) inoculated plants. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315 or in S3 Table.

(TIFF)

pbio.3003358.s004.tiff^{(1.5MB, tiff)}

S5 Fig. Untargeted metabolomics on rhizosphere metabolites.

NMDS plots with a 95% confidence ellipse for 833 filter features for individual laboratories A–E and all combined. Different colors show treatments: Axenic (gray), SynCom16 (blue), and SynCom17 (orange), while shapes indicate laboratories in the combined plot. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s005.tiff^{(1.2MB, tiff)}

S6 Fig. Statistical analysis of 16S rRNA sequencing data.

(a) ANOVA p-value of effects for root and media. Each point is a SynCom member with blue indicating significant values (red line p = 0.05). (b) ANOVA source of variation for root and media calculated from ratio of squares (SS) for each effect and total sum of squares (TSS). (c) Coefficient of variation (CV = σ/μ) distribution (median in red) for treatments within (blue) and across (gray) labs for root and media. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s006.tiff^{(1.8MB, tiff)}

S7 Fig. Statistical analysis of 60 identified rhizosphere metabolites.

(a) ANOVA p-value of effects for metabolite peak heights. Each point is a metabolite with blue indicating significant values (red line p = 0.05). (b) Number of statistically different metabolites (p < 0.05) in lab pairwise comparisons. (c) ANOVA source of variation for root and media calculated from ratio of squares (SS) for each effect and total sum of squares (TSS). (d) Coefficient of variation (CV = σ/μ) distribution (median in red) for treatments within (blue) and across (gray) labs. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s007.tiff^{(2MB, tiff)}

S8 Fig. Comparative genomics of bacterial isolates.

Bar graphs show genome characteristics (from left: the abundance of bases, coding, G + C bases, total genes, and KEGG pathway genes. The search for acid resistance system genes included GAD (EC 4.1.1.15 glutamate decarboxylase), AdiA (EC 4.1.1.19 arginine decarboxylase), glsA (EC 3.5.1.2 glutaminase), GlnHPQ (glutamine ABC transporter), GltIJKL (glutamate/aspartate ABC transporter), HisPMQ-ArgT (arginine/ornithine ABC transporter). The heat map shows normalized gene abundance for selected KEGG pathways. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s008.tiff^{(2.7MB, tiff)}

S9 Fig. Fluorescent microscopy in EcoFAB 2.0 conducted on plants grown in the lab A.

We inoculated RFP-expressing Paraburkholderia sp. OAS925 into the B. distachyon rhizosphere in EcoFAB 2.0 devices on the day of transfer (DAT) for the seedling. The plots show (a) uninoculated plant control (1 DAT) and medium-inoculated plants (OD₆₀₀ 0.01) at (b) 1 and (c) 3 days after inoculation (DAI). EcoFAB 2.0 root scans indicate the locations for microscopy. The inset microscopy images show merged TxRed and bright-field (BF) channels. The microscopy images can be found at https://doi.org/10.6084/m9.figshare.26449852.

(TIFF)

pbio.3003358.s009.tiff^{(8.2MB, tiff)}

S10 Fig. Assessment of growth and biofilm formation for bacterial isolates in vitro by the lab A.

(Top) Biofilm was measured with crystal violet (CV) staining at OD_550, and (Bottom) isolate growth was assessed based on OD₆₀₀ values of isolates on the defined NLDM liquid medium for 3 days. The horizontal dotted line indicates the mean value of sterile medium negative control (NC). Different letters indicate statistically significant differences at p < 0.05, One-way ANOVA with Tukey’s test, n = 4–5 for isolates and n = 11 for NC. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s010.tiff^{(933.3KB, tiff)}

S11 Fig. Motility assays in the lab A.

(a) The initial screen for swimming motility across bacterial isolates was measured 24 and 45 h after inoculation. (b) Phenotypes of the most motile strains at 45 h since inoculation. (c) pH effects on the motility ring diameter of isolates with bulls-eye colony morphology. Images of motility assays with bacterial isolates can be found at https://doi.org/10.6084/m9.figshare.26457928. The data underlying this figure can be found at https://doi.org/10.6084/m9.figshare.26401315.

(TIFF)

pbio.3003358.s011.tiff^{(5MB, tiff)}

S1 Table. Chamber settings and models and experiment timing.

(XLSX)

pbio.3003358.s012.xlsx^{(71.1KB, xlsx)}

S2 Table. SynCom members overview, genome information, and their OD₆₀₀ to CFU conversion ratios.

(XLSX)

pbio.3003358.s013.xlsx^{(15.7KB, xlsx)}

S3 Table. 16S rRNA sequencing read counts in media and root samples.

Samples labeled “none” are Axenic plant controls, “Gly” are glycerol stocks of inoculum.

(XLSX)

pbio.3003358.s014.xlsx^{(51.4KB, xlsx)}

S4 Table. BLAST search results for unknown reads in selected media samples.

(XLSX)

pbio.3003358.s015.xlsx^{(13.1KB, xlsx)}

S5 Table. LC–MS/MS parameters.

(XLSX)

pbio.3003358.s016.xlsx^{(11.8KB, xlsx)}

S6 Table. Metabolite identification and intensity.

Sample names follow code S-L-T, where S is sample (ExCtrl = extraction control, RtExu = root exudate sample, TxCtrl = technical control); L is the laboratory (A–E letters for each lab); T is the treatment (Axenic, SynCom16 or SynCom17). The prefix PkHt = Peak height.

(XLSX)

pbio.3003358.s017.xlsx^{(175.9KB, xlsx)}

S7 Table. ANOVA results for 16S rRNA sequencing data and metabolite intensity.

(XLSX)

pbio.3003358.s018.xlsx^{(53.3KB, xlsx)}

S8 Table. Pairwise results for metabolites.

(XLSX)

pbio.3003358.s019.xlsx^{(213.9KB, xlsx)}

S9 Table. Comparison of KEGG modules between SynCom16 vs. Paraburkholderia sp. OAS925.

(XLSX)

pbio.3003358.s020.xlsx^{(35.6KB, xlsx)}

S1 Video. Fluorescent microscopy at 1 DAI.

Z-Stack shows RFP-Paraburkholderia motility and colonization on B. distachyon roots at 1 DAI in EcoFAB 2.0.

(MP4)

Download video file^{(8.8MB, mp4)}

S2 Video. Fluorescent microscopy at 3 DAI.

Root colonization by RFP-Paraburkholderia at 3 DAI in EcoFAB 2.0. Merged bright-field and TxRed channels followed by TxRed footage.

(MP4)

Download video file^{(11.5MB, mp4)}

S1 File. MZMine 3 settings.

(ZIP)

pbio.3003358.s023.zip^{(3.8KB, zip)}

Acknowledgments

We would like to thank Julio Corral and Kaosio Saephan for coordinating the shipment and receipt of samples and Chips Hoai for guiding us regarding regulations related to the import and export of samples. We thank Diana Dresbach for technical help with the trial at the Max Planck Institute.

Data analysis utilized resources from the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility (Contract No. DE-AC02-05CH11231). A portion of these data was produced by the US Department of Energy Joint Genome Institute (https://ror.org/04xm1d337; operated under Contract No. DE-AC02-05CH11231) as part of projects 10.46936/10.25585/60001370 and 10.46936/10.25585/60001258.

Abbreviations

:

ANOVA: analysis of variation
CFU: colony-forming unit
CV: coefficient of variation
DAI: days after inoculation
DAP: diaminopimelic acid
DSMZ: Leibniz-Institute DSMZ-German Collection of Microorganisms and Cell Cultures
EcoFAB: fabricated ecosystem
IMG/M: Integrated Microbial Genomes and Microbiomes
KEGG: Kyoto Encyclopedia of Genes and Genomes
LB: Luria–Bertani
LBNL: Lawrence Berkeley National Laboratory
LC-MS/MS: liquid chromatography-tandem mass spectrometry
MS: Murashige and Skoog
NMDC: National Microbiome Data Collaborative
NMDS: Non-metric Multidimensional Scaling
OD600: optical density at 600 nm
OTU: operational taxonomic unit
RFP: red fluorescent protein
RT: retention time
SynCom: synthetic community
T3SS: Type 3 Secretion System

Data Availability

All of the links to the protocols and data as well as the study metadata are available via NMDC at https://data.microbiomedata.org/details/study/nmdc:sty-11-ev70y104. The 16S rRNA amplicon sequencing data are available via NCBI (https://www.ncbi.nlm.nih.gov/) as BioProject PRJNA1151037. All raw data, including plant phenotypes, sterility tests, metabolite identifications, in vitro assays, and code are available via Figshare at https://doi.org/10.6084/m9.figshare.c.7373842. The untargeted metabolomics outputs (HILIC-pos) with features annotations and .mzml files are available via GNPS2 at https://gnps2.org/status?task=2ccbf82840724c99a2acc2c9e512a302. Raw LC–MS/MS files are available in .raw format at MassIVE (https://massive.ucsd.edu/) under ID number MSV000095476 or via https://doi.org/10.25345/C5Q23RB6B. The protocol is available at protocols.io via https://dx.doi.org/10.17504/protocols.io.kxygxyydkl8j/v1. The annotated bacterial genomes can be accessed via IMG/M (https://img.jgi.doe.gov/) by searching for either the isolate name, taxon ID, or GOLD Project ID; on Hugging Face (https://doi.org/10.57967/hf/5885); or by using the links in S2 Table.

Funding Statement

We gratefully acknowledge the financial support from the U.S. Department of Energy (DOE) Office of Science, Office of Biological and Environmental Research for support of the Trial Ecosystem Advancement for Microbiome Science (TEAMS) led by LBNL, Award DE-SC0021234 led by UC San Diego, and m-CAFEs Microbial Community Analysis & Functional Evaluation in Soils (m-CAFEs@lbl.gov), a Science Focus Area led by Lawrence Berkeley National Laboratory where all of the LBNL led projects are under Contract No. DE-AC02-05CH11231. J.C. and M.W. acknowledge support from the University of Melbourne Botany Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Zengler K, Hofmockel K, Baliga NS, Behie SW, Bernstein HC, Brown JB, et al. EcoFABs: advancing microbiome science through standardized fabricated ecosystems. Nat Methods. 2019;16(7):567–71. doi: 10.1038/s41592-019-0465-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.O’Toole GA. We have a community problem. J Bacteriol. 2024;206(4):e0007324. doi: 10.1128/jb.00073-24 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Vorholt JA, Vogel C, Carlström CI, Müller DB. Establishing causality: opportunities of synthetic communities for plant microbiome research. Cell Host Microbe. 2017;22(2):142–55. doi: 10.1016/j.chom.2017.07.004 [DOI] [PubMed] [Google Scholar]
4.Marín O, González B, Poupin MJ. From microbial dynamics to functionality in the rhizosphere: a systematic review of the opportunities with synthetic microbial communities. Front Plant Sci. 2021;12:650609. doi: 10.3389/fpls.2021.650609 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Salas-González I, Reyt G, Flis P, Custódio V, Gopaulchan D, Bakhoum N, et al. Coordination between microbiota and root endodermis supports plant mineral nutrient homeostasis. Science. 2021;371(6525):eabd0695. doi: 10.1126/science.abd0695 [DOI] [PubMed] [Google Scholar]
6.Finkel OM, Salas-González I, Castrillo G, Conway JM, Law TF, Teixeira PJPL, et al. A single bacterial genus maintains root growth in a complex microbiome. Nature. 2020;587(7832):103–8. doi: 10.1038/s41586-020-2778-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Vandenkoornhuyse P, Quaiser A, Duhamel M, Le Van A, Dufresne A. The importance of the microbiome of the plant holobiont. New Phytol. 2015;206(4):1196–206. doi: 10.1111/nph.13312 [DOI] [PubMed] [Google Scholar]
8.Northen TR, Kleiner M, Torres M, Kovács ÁT, Nicolaisen MH, Krzyżanowska DM. Community standards and future opportunities for synthetic communities in plant-microbiota research. Nat Microbiol. 2024;9(11):2774–84. doi: 10.1038/s41564-024-01833-4 [DOI] [PubMed] [Google Scholar]
9.Coker J, Zhalnina K, Marotz C, Thiruppathy D, Tjuanta M, D’Elia G, et al. A reproducible and tunable synthetic soil microbial community provides new insights into microbial ecology. mSystems. 2022;7(6):e0095122. doi: 10.1128/msystems.00951-22 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Sasse J, Kant J, Cole BJ, Klein AP, Arsova B, Schlaepfer P, et al. Multilab EcoFAB study shows highly reproducible physiology and depletion of soil metabolites by a model grass. New Phytol. 2019;222(2):1149–60. doi: 10.1111/nph.15662 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Lin H-H, Torres M, Adams CA, Andeer PF, Owens TK, Zhalnina K, et al. Impact of inoculation practices on microbiota assembly and community stability in a fabricated ecosystem. Phytobiomes J. 2024;8(2):155–67. doi: 10.1094/pbiomes-06-23-0050-r [DOI] [Google Scholar]
12.Novak V, Andeer PF, Bowen BP, Ding Y, Zhalnina K, Hofmockel KS, et al. Reproducible growth of Brachypodium in EcoFAB 2.0 reveals that nitrogen form and starvation modulate root exudation. Sci Adv. 2024;10(1):eadg7888. doi: 10.1126/sciadv.adg7888 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.McGrath TF, Haughey SA, Islam M, Elliott CT, Collaborators. The potential of handheld near infrared spectroscopy to detect food adulteration: results of a global, multi-instrument inter-laboratory study. Food Chem. 2021;353:128718. doi: 10.1016/j.foodchem.2020.128718 [DOI] [PubMed] [Google Scholar]
14.Thompson M, Ellison SLR, Wood R. The International Harmonized Protocol for the proficiency testing of analytical chemistry laboratories (IUPAC Technical Report). Pure Appl Chem. 2006;78(1):145–96. doi: 10.1351/pac200678010145 [DOI] [Google Scholar]
15.Novak V, King E, Andeer P, Vogel J, t Northen T. Use of EcoFAB 2.0 for reproducible plant-microbe interaction experiments v1. 2024. Available from: doi: 10.17504/protocols.io.kxygxyydkl8j/v1 [DOI] [Google Scholar]
16.Bowen B. SynCom17_EcoFab. Hugging Face. 2025. doi: 10.57967/hf/5885 [DOI] [Google Scholar]
17.Zhalnina K, Louie KB, Hao Z, Mansoori N, da Rocha UN, Shi S, et al. Dynamic root exudate chemistry and microbial substrate preferences drive patterns in rhizosphere microbial community assembly. Nat Microbiol. 2018;3(4):470–80. doi: 10.1038/s41564-018-0129-3 [DOI] [PubMed] [Google Scholar]
18.Korenblum E, Dong Y, Szymanski J, Panda S, Jozwiak A, Massalha H, et al. Rhizosphere microbiome mediates systemic root metabolite exudation by root-to-root signaling. Proc Natl Acad Sci U S A. 2020;117(7):3874–83. doi: 10.1073/pnas.1912130117 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Allison SD, Vitousek PM. Responses of extracellular enzymes to simple and complex nutrient inputs. Soil Biol Biochemi. 2005;37(5):937–44. doi: 10.1016/j.soilbio.2004.09.014 [DOI] [Google Scholar]
20.de Raad M, Li YV, Kuehl JV, Andeer PF, Kosina SM, Hendrickson A, et al. A defined medium for cultivation and exometabolite profiling of soil bacteria. Front Microbiol. 2022;13:855331. doi: 10.3389/fmicb.2022.855331 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Schloss PD. Identifying and overcoming threats to reproducibility, replicability, robustness, and generalizability in microbiome research. mBio. 2018;9(3):e00525–18. doi: 10.1128/mBio.00525-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Meiser J, Weindl D, Hiller K. Complexity of dopamine metabolism. Cell Commun Signal. 2013;11(1):34. doi: 10.1186/1478-811X-11-34 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Umek N, Geršak B, Vintar N, Šoštarič M, Mavri J. Dopamine autoxidation is controlled by acidic pH. Front Mol Neurosci. 2018;11:467. doi: 10.3389/fnmol.2018.00467 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Nuccio EE, Starr E, Karaoz U, Brodie EL, Zhou J, Tringe SG, et al. Niche differentiation is spatially and temporally regulated in the rhizosphere. ISME J. 2020;14(4):999–1014. doi: 10.1038/s41396-019-0582-x [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Fierer N. Embracing the unknown: disentangling the complexities of the soil microbiome. Nat Rev Microbiol. 2017;15(10):579–90. doi: 10.1038/nrmicro.2017.87 [DOI] [PubMed] [Google Scholar]
26.Debray R, Herbert RA, Jaffe AL, Crits-Christoph A, Power ME, Koskella B. Priority effects in microbiome assembly. Nat Rev Microbiol. 2022;20(2):109–21. doi: 10.1038/s41579-021-00604-w [DOI] [PubMed] [Google Scholar]
27.Carlström CI, Field CM, Bortfeld-Miller M, Müller B, Sunagawa S, Vorholt JA. Synthetic microbiota reveal priority effects and keystone strains in the Arabidopsis phyllosphere. Nat Ecol Evol. 2019;3(10):1445–54. doi: 10.1038/s41559-019-0994-z [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Kuzyakov Y, Razavi BS. Rhizosphere size and shape: temporal dynamics and spatial stationarity. Soil Biol Biochem. 2019;135:343–60. doi: 10.1016/j.soilbio.2019.05.011 [DOI] [Google Scholar]
29.Lu P, Ma D, Chen Y, Guo Y, Chen G-Q, Deng H, et al. L-glutamine provides acid resistance for Escherichia coli through enzymatic release of ammonia. Cell Res. 2013;23(5):635–44. doi: 10.1038/cr.2013.13 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Minamino T, Imae Y, Oosawa F, Kobayashi Y, Oosawa K. Effect of intracellular pH on rotational speed of bacterial flagellar motors. J Bacteriol. 2003;185(4):1190–4. doi: 10.1128/JB.185.4.1190-1194.2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Cole BJ, Feltcher ME, Waters RJ, Wetmore KM, Mucyn TS, Ryan EM, et al. Genome-wide identification of bacterial plant colonization genes. PLoS Biol. 2017;15(9):e2002860. doi: 10.1371/journal.pbio.2002860 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Hafner K, Welt B, Pelletier W. Dry ice sublimation performance as affected by binding agent, density, and age. Engineering. 2023. [Google Scholar]
33.Teytelman L, Stoliartchouk A, Kindler L, Hurwitz BL. Protocols.io: virtual communities for protocol development and discussion. PLoS Biol. 2016;14(8):e1002538. doi: 10.1371/journal.pbio.1002538 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Eloe-Fadrosh EA, Ahmed F, Babinski M, Baumes J, Borkum M, Bramer L, et al. The National Microbiome Data Collaborative Data Portal: an integrated multi-omics microbiome data resource. NAT. 2022. doi: 10.1093/nar/gkab990 [DOI] [Google Scholar]
35.Sinha R, Abu-Ali G, Vogtmann E, Fodor AA, Ren B, Amir A, et al. Assessment of variation in microbial community amplicon sequencing by the Microbiome Quality Control (MBQC) project consortium. Nat Biotechnol. 2017;35(11):1077–86. doi: 10.1038/nbt.3981 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Castrillo G, Teixeira PJPL, Paredes SH, Law TF, de Lorenzo L, Feltcher ME, et al. Root microbiota drive direct integration of phosphate stress and immunity. Nature. 2017;543(7646):513–8. doi: 10.1038/nature21417 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Durán P, Thiergart T, Garrido-Oter R, Agler M, Kemen E, Schulze-Lefert P, et al. Microbial interkingdom interactions in roots promote Arabidopsis survival. Cell. 2018;175(4):973-983.e14. doi: 10.1016/j.cell.2018.10.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Sordo Z, Andeer P, Sethian J, Northen T, Ushizima D. RhizoNet segments plant roots to assess biomass and growth for enabling self-driving labs. Sci Rep. 2024;14(1):12907. doi: 10.1038/s41598-024-63497-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Lobet G, Pagès L, Draye X. A novel image-analysis toolbox enabling quantitative analysis of root system architecture. Plant Physiol. 2011;157(1):29–39. doi: 10.1104/pp.111.179895 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Schmid R, Heuckeroth S, Korf A, Smirnov A, Myers O, Dyrlund TS, et al. Integrative analysis of multimodal mass spectrometry data in MZmine 3. Nat Biotechnol. 2023;41(4):447–9. doi: 10.1038/s41587-023-01690-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, et al. Sharing and community curation of mass spectrometry data with Global Natural Products Social Molecular Networking. Nat Biotechnol. 2016;34(8):828–37. doi: 10.1038/nbt.3597 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Bowen BP, Northen TR. Dealing with the unknown: metabolomics and metabolite atlases. J Am Soc Mass Spectrom. 2010;21(9):1471–6. doi: 10.1016/j.jasms.2010.04.003 [DOI] [PubMed] [Google Scholar]
43.Shaffer JP, Nothias L-F, Thompson LR, Sanders JG, Salido RA, Couvillion SP, et al. Standardized multi-omics of Earth’s microbiomes reveals microbial and metabolite diversity. Nat Microbiol. 2022;7(12):2128–50. doi: 10.1038/s41564-022-01266-x [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460–1. doi: 10.1093/bioinformatics/btq461 [DOI] [PubMed] [Google Scholar]
45.Haney EF, Trimble MJ, Hancock REW. Microtiter plate assays to assess antibiofilm activity against bacteria. Nat Protoc. 2021;16(5):2615–32. doi: 10.1038/s41596-021-00515-3 [DOI] [PubMed] [Google Scholar]
46.Mukherjee S, Stamatis D, Li CT, Ovchinnikova G, Bertsch J, Sundaramurthi JC, et al. Twenty-five years of Genomes OnLine Database (GOLD): data updates and new features in v.9. Nucleic Acids Res. 2023;51(D1):D957–63. doi: 10.1093/nar/gkac974 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Chen I-MA, Chu K, Palaniappan K, Ratner A, Huang J, Huntemann M, et al. The IMG/M data management and analysis system v.7: content updates and new features. Nucleic Acids Res. 2023;51(D1):D723–32. doi: 10.1093/nar/gkac976 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Chen I-MA, Chu K, Palaniappan K, Ratner A, Huang J, Huntemann M, et al. The IMG/M data management and analysis system v.6.0: new tools and advanced capabilities. Nucleic Acids Res. 2021;49(D1):D751–63. doi: 10.1093/nar/gkaa939 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Parks DH, Chuvochina M, Rinke C, Mussig AJ, Chaumeil P-A, Hugenholtz P. GTDB: an ongoing census of bacterial and archaeal diversity through a phylogenetically consistent, rank normalized and complete genome-based taxonomy. Nucleic Acids Res. 2022;50: D785–D794. doi: 10.1093/nar/gkab776 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Price MN, Dehal PS, Arkin AP. FastTree 2—approximately maximum-likelihood trees for large alignments. PLoS One. 2010;5(3):e9490. doi: 10.1371/journal.pone.0009490 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 2021;49(W1):W293–6. doi: 10.1093/nar/gkab301 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Pearson AN, Thompson MG, Kirkpatrick LD, Ho C, Vuu KM, Waldburger LM, et al. The pGinger family of expression plasmids. Microbiol Spectr. 2023;11(3):e0037323. doi: 10.1128/spectrum.00373-23 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Warnes GR, Bolker B, Bonebakker L, Gentleman R, Huber W, Liaw A, et al. gplots: Various R Programming Tools for Plotting Data. R package version 3.1.3.1. 2024 [cited 31 Mar 2024]. Available from: https://CRAN.R-project.org/package=gplots
54.Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community Ecology Package. 2020. Available from: http://CRAN.R-project.org/package=vegan

PLoS Biol. doi: 10.1371/journal.pbio.3003358.r001

Decision Letter 0

Melissa Vazquez Hernandez

17 Jan 2025

Dear Trent,

Thank you for submitting your manuscript entitled "Multi-laboratory Study Establishes Reproducible Methods for Plant-Microbiome Research in Fabricated Ecosystems" for consideration as a Methods and Resources by PLOS Biology.

Your manuscript has now been evaluated by the PLOS Biology editorial staff, as well as by an academic editor with relevant expertise, and I am writing to let you know that we would like to send your submission out for external peer review. We would like to suggest to change the type of article to Meta-Research Article since we think might fit better. Meta-Research Articles examine how biological research is designed, carried out, communicated and evaluated; We welcome both exploratory and confirmatory research that has the potential to drive change in research and evaluation practices in the life sciences and beyond.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire. If you agree with the change in Article type, please, when adding the rest of the metadata choose "Meta-Research Article".

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. After your manuscript has passed the checks it will be sent out for review. To provide the metadata for your submission, please Login to Editorial Manager (https://www.editorialmanager.com/pbiology) within two working days, i.e. by Jan 19 2025 11:59PM.

If your manuscript has been previously peer-reviewed at another journal, PLOS Biology is willing to work with those reviews in order to avoid re-starting the process. Submission of the previous reviews is entirely optional and our ability to use them effectively will depend on the willingness of the previous journal to confirm the content of the reports and share the reviewer identities. Please note that we reserve the right to invite additional reviewers if we consider that additional/independent reviewers are needed, although we aim to avoid this as far as possible. In our experience, working with previous reviews does save time.

If you would like us to consider previous reviewer reports, please edit your cover letter to let us know and include the name of the journal where the work was previously considered and the manuscript ID it was given. In addition, please upload a response to the reviews as a 'Prior Peer Review' file type, which should include the reports in full and a point-by-point reply detailing how you have or plan to address the reviewers' concerns.

During the process of completing your manuscript submission, you will be invited to opt-in to posting your pre-review manuscript as a bioRxiv preprint. Visit http://journals.plos.org/plosbiology/s/preprints for full details. If you consent to posting your current manuscript as a preprint, please upload a single Preprint PDF.

Feel free to email us at plosbiology@plos.org if you have any queries relating to your submission.

Have a nice weekend,

Melissa

Melissa Vazquez Hernandez, Ph.D.

Associate Editor

PLOS Biology

mvazquezhernandez@plos.org

PLoS Biol. doi: 10.1371/journal.pbio.3003358.r002

Decision Letter 1

Melissa Vazquez Hernandez

26 Mar 2025

Dear Trent,

Thank you for your patience while your manuscript "Multi-laboratory Study Establishes Reproducible Methods for Plant-Microbiome Research in Fabricated Ecosystems" was peer-reviewed at PLOS Biology. It has now been evaluated by the PLOS Biology editors, an Academic Editor with relevant expertise, and by two independent reviewers. First of all I would like to apologize for the extremely long delay on giving you a decision, this is really out of the normal.

In light of the reviews, which you will find at the end of this email, we would like to invite you to revise the work to thoroughly address the reviewers' reports. As you will see below, the reviewers are positive about the relevance and novelty of the study, yet some concerns have raised during revision. Reviewer 1 mentions that you might be missing additional analysis on the differences that exist between the labs. The reviewer suggests to model microbiome data including the lab as a factor, consider how uneven the inoculum seems to be, and clarify several points on their experiments. While most of the comments from Reviewer 2 could be addressed with further discussion and clarification in the text, the reviewer mentions that perhaps you should re-do the experiments to ensure absence of contamination. Additionally, the reviewer mentions some experiments to answer e.g. how does a single strain Paraburkholderia sp. OAS925 affects B. distachyon growth.

IMPORTANT: after discussion with the Academic Editor and the reviewers, while we think reanalysis of the data might be necessary, we do not expect that you re-do any experiment as contamination tends to be a reality in such studies.

Given the extent of revision needed, we cannot make a decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is likely to be sent for further evaluation by all or a subset of the reviewers.

In addition to these revisions, you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests shortly.

We expect to receive your revised manuscript within 3 months. Please email us (plosbiology@plos.org) if you have any questions or concerns, or would like to request an extension.

At this stage, your manuscript remains formally under active consideration at our journal; please notify us by email if you do not intend to submit a revision so that we may withdraw it.

**IMPORTANT - SUBMITTING YOUR REVISION**

Your revisions should address the specific points made by each reviewer. Please submit the following files along with your revised manuscript:

1. A 'Response to Reviewers' file - this should detail your responses to the editorial requests, present a point-by-point response to all of the reviewers' comments, and indicate the changes made to the manuscript.

*NOTE: In your point-by-point response to the reviewers, please provide the full context of each review. Do not selectively quote paragraphs or sentences to reply to. The entire set of reviewer comments should be present in full and each specific point should be responded to individually, point by point.

You should also cite any additional relevant literature that has been published since the original submission and mention any additional citations in your response.

2. In addition to a clean copy of the manuscript, please also upload a 'track-changes' version of your manuscript that specifies the edits made. This should be uploaded as a "Revised Article with Changes Highlighted" file type.

*Re-submission Checklist*

When you are ready to resubmit your revised manuscript, please refer to this re-submission checklist: https://plos.io/Biology_Checklist

To submit a revised version of your manuscript, please go to https://www.editorialmanager.com/pbiology/ and log in as an Author. Click the link labelled 'Submissions Needing Revision' where you will find your submission record.

Please make sure to read the following important policies and guidelines while preparing your revision:

*Published Peer Review*

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. Please see here for more details:

https://blogs.plos.org/plos/2019/05/plos-journals-now-open-for-published-peer-review/

*PLOS Data Policy*

Please note that as a condition of publication PLOS' data policy (http://journals.plos.org/plosbiology/s/data-availability) requires that you make available all data used to draw the conclusions arrived at in your manuscript. If you have not already done so, you must include any data used in your manuscript either in appropriate repositories, within the body of the manuscript, or as supporting information (N.B. this includes any numerical values that were used to generate graphs, histograms etc.). For an example see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5

*Blot and Gel Data Policy*

We require the original, uncropped and minimally adjusted images supporting all blot and gel results reported in an article's figures or Supporting Information files. We will require these files before a manuscript can be accepted so please prepare them now, if you have not already uploaded them. Please carefully read our guidelines for how to prepare and upload this data: https://journals.plos.org/plosbiology/s/figures#loc-blot-and-gel-reporting-requirements

*Protocols deposition*

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Thank you again for your submission to our journal. We hope that our editorial process has been constructive thus far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Melissa

Melissa Vazquez Hernandez, Ph.D.

Associate Editor

PLOS Biology

mvazquezhernandez@plos.org

------------------------------------

REVIEWERS' COMMENTS:

------------------------------------

Reviewer #1:

This manuscript addresses the important and rarely-investigated issue of replicability of results between labs, specifically applied here to microbiome research. The organizers got together 5 participating labs and distributed an axenic growth chamber, pre-mixed microbial communities, plant seeds, and specific instructions to attempt to reproduce the same experiment as faithfully as possible. Much, but not all, of the experiment replicated. The paper is interesting, but should do some additional analyses on the results, particularly related to the differences that do exist between labs. The authors focused mainly on how Paraburkholderia completely overtook the community in all labs. To some extent, the detailed look at Paraburkholderia feels like a bit of a distraction from the main interesting points about reproducibility, because it is SOOO dominant. Focusing on this obscures other instructive differences, which I think should be more central. More specific comments follow, not in any particular order.

>> The authors should model microbiome data including lab as a factor. How much of the variation in community in SynCom 16 is due to lab? How much remains unexplained? For community 17, after removing Paraburkholderia reads, what is left? There are unsatisfyingly few stats.

>> A couple things are curious about the inoculum. First, it appears quite uneven - I suppose the 16 or 17 members were intended to be mixed quite evenly. Can this be further explained? Second, the inoculum seems to bear nearly no resemblance to the ultimate community. Even Paraburkholderia seems not to be present in any substantial amount in the inoculum. Can the authors please present a table of percentages (in supplemental would be OK) that lists the median % of each community member in the inoculum and also in the ultimate community?

>>Related to the inoculum, it is odd that the inoculum in Figure 3 (plants) appears different than the inoculum in figure S3 (media). Were the media and plant inoculations NOT done simultaneously, such that independent inocula batches were thawed for the media?

>>Particularly the media in Fig. S3, there are visible bars of "unknown" sequences colored black. What is the diversity of "unknown" in this? For example, if the "unknown" is all the same thing, isn't that a clear case of contamination? And if it's a mix of different things, how to explain it?

>> it is odd that the inoculum in Figure 3 (plants) appears different than the inoculum in figure S3 (media). Were the media and plant inoculations NOT done simultaneously, such that independent inocula batches were thawed for the media?

>> For some analysis (eg microscopy) was done in the main host lab. I think it is worth identifying which of the labs that was. Those analysis could mention, for example, that microscropy was conducted on plants grown in Lab A.

>> In Figure S1, each column (group of measurements of the same phenotype) should have the same Y axis so that the plots can be compared. And I think it's further preferable that columns and rows are switched, such that one can look across from left to right and see the shoot dry weight in each lab, for example.

>>Likewise in Figure S2, the Y axes should be made comparable. The temperature side is OK, but the light levels are highly variable across labs and this should be immediately apparent looking at the graphs, but it's somewhat hidden due to the variable axes.

>>Fig. 2A is confusing. The barplots look like this is a continuous distribution, but as I understand it, if a device was sterile, it gets a grey box, and if it's contaminated, it gets a black box. It would be clearer if the boxes were separated so they could be seen as discrete data points, or even I think it would be clearer if the Y axis showed # of contaminated boxes and most are 0 and those that got contaminated go to 1. For example. But it's confusing as is.

>> in Fig 2C, there are some differences between the labs in terms of measurement times. Lab B took a measurement at 4 days? Lab E did 0 days but skipped 14 days? What happened here, a miscommunication or technical failure or something?

>> Figure S4, putting each lab with a totally different NMDS2 plot is difficult (for me) to interpret. Is not it easier and more informative to simply use the combined lab plot to generate the spacings for the NMDS and simply erase the points from the other labs so that points from a single lab can be shown individually?

>> In the methods, please include the min/max/median sequencing depth for the microbiome samples.

>>Line 276.. the authors say the ecofab devices are "free". Surely this can't be true, or must have some limits. Please revise to clarify. It means free if one is part of a JGI user program? Is that limited to USA? Grant recipients of some form? Etc.

------------------------------------

Reviewer #2 (Mengcen Wang):

This study addresses a critical challenge in microbiome research, inter-laboratory reproducibility, by assessing the consistency of synthetic community assembly experiments across multiple laboratories. The authors leverage ecosystems involving Brachypodium distachyon, two different synthetic bacterial communities, and sterile EcoFAB 2.0 devices, providing valuable insights into inoculum-dependent microbiome shifts and plant responses. While the study is generally well-structured and presents important findings, some areas require further clarification and improvement:

1. It was indicated that the root system development was analyzed using RhizoNet (Lab B-E) and ImageJ (Lab A). Why did these five laboratories not standardize their analytical methods to minimize variability and improve reproducibility? Furthermore, it would be beneficial to include representative images of root systems.

2. In line135-139, the uninoculated EcoFAB 2.0 sterility tests across laboratories A-E shown that the treatments of laboratory D in SynCom17 and laboratory B in medium-only control were contaminated. It would be more appropriate to redo the experiments to ensure the absence of contamination and further validate the reliability of the results.

3. The SynCom17, which contained the dominating bacteria Paraburkholderia sp. OAS925 could reduce root growth of B. distachyon. How about the effects of the single strain Paraburkholderia sp. OAS925 on B. distachyon growth?

4. L209-210 what does " its potential to outgrow competitors when cultured with common soil metabolites in the NLDM " mean?

5. L226-L228, "This finding is consistent with a previous study showing Paraburkholderia sp. OAS925 dominance in the B. distachyon root and rhizosphere microbiota and decreased fresh root biomass." However, in Figure 2b, the shoot fresh weight (FW) of plants treated with both SynCom 16 and SynCom 17 was significantly reduced compared to axenic conditions. Moreover, no significant difference was observed between SynCom 16 and SynCom 17 treatments, suggesting that the SynCom 16 community, even without the addition of Paraburkholderia sp. OAS925, is sufficient to reduce shoot FW levels. Therefore, this does not conclusively demonstrate that the reduction in shoot FW under SynCom 17 treatment is specifically driven by Paraburkholderia sp. OAS925.

6. In L303, the manuscript does not clearly explain the rationale behind the 1:1 ratio of strains in the SynComs. It is recommended to provide additional explanation to enhance the scientific rigor and credibility of the study.

7. In L362, the authors indicate that the hydroponic medium sample was used as the root exudate for metabolomic analysis. But how did they rule out potential interference from bacterial metabolites in the medium?

8. The colonization of bacteria in B. distachyon roots can be influenced by various external factors. The assessment of bacterial motility on liquid soft agar only reflects swimming ability in liquid culture media and does not provide definitive evidence that Paraburkholderia sp. OAS925 can outcompete other members during SynCom colonization.

9. The analysis of microbial community and metabolome data is crucial for the research conclusions. However, it is not clear in the paper whether significance analysis was performed on this part of the data and what methods were used. If there is relevant analysis, please provide a complete description.

10. It is recommended to present the analysis conclusions of some data in the supplementary information to enrich the content of the main text.

PLoS Biol. 2025 Sep 8;23(9):e3003358. doi: 10.1371/journal.pbio.3003358.r003

Author response to Decision Letter 2

23 Jun 2025

Attachment

Submitted filename: Responses.pdf

pbio.3003358.s026.pdf^{(9.7MB, pdf)}

PLoS Biol. doi: 10.1371/journal.pbio.3003358.r004

Decision Letter 2

Melissa Vazquez Hernandez

25 Jul 2025

Dear Trent,

I hope you are doing great. Thank you for your patience while we considered your revised manuscript "Multi-laboratory Study Establishes Reproducible Methods for Plant-Microbiome Research in Fabricated Ecosystems" for publication as a Meta-Research Article at PLOS Biology. This revised version of your manuscript has been evaluated by the PLOS Biology editors, and the original reviewers.

Based on the reviews, we are likely to accept this manuscript for publication, provided you satisfactorily address the remaining editorial points. Please be aware that Academic Editor was unavailable and there might be a possibility that they might have an additional request, which I would communicate with you if it were the case. Please make sure to address the following data and other policy-related requests.

a) We routinely suggest changes to titles to ensure maximum accessibility for a broad, non-specialist readership, and to ensure they reflect the contents of the paper. In this case, we would suggest a minor edit to the title, as follows. Please ensure you change both the manuscript file and the online submission system, as they need to match for final acceptance:

"Protocols, benchmarking datasets and best practices for reproducible research on plant-microbiome interactions"

b) You may be aware of the PLOS Data Policy, which requires that all data be made available without restriction: http://journals.plos.org/plosbiology/s/data-availability. For more information, please also see this editorial: http://dx.doi.org/10.1371/journal.pbio.1001797

Please supply the numerical values either in the a supplementary file or as a permanent DOI’d deposition for the following figures:

Figure 2bc, 3, 4, S1, S2ab, S4ab, S5, S6bc, S7bcd, S8, S9, S11ac

NOTE: the numerical data provided should include all replicates AND the way in which the plotted mean and errors were derived (it should not present only the mean/average values).

*I am aware that there is some raw data in Figshare but it is not clear if it belongs to these figures. If this is the case, I need you to provide the necessary information to be able to replicate each figure. However, it might be easier to just provide an excel file with the raw data for each figure.

c) Please cite the location of the data clearly in all relevant main and supplementary Figure legends, e.g. “The data underlying this Figure can be found in S1 Data” or “The data underlying this Figure can be found in https://doi.org/10.5281/zenodo.XXXXX”

d) Could you please confirm that the repositories you used are publicly available?

e) Please ensure that your Data Statement in the submission system accurately describes where your data can be found and is in final format, as it will be published as written there

f) Per journal policy, if you have generated any custom code during the course of this investigation, please make it available without restrictions. Please ensure that the code is sufficiently well documented and reusable, and that your Data Statement in the Editorial Manager submission system accurately describes where your code can be found.

As you address these items, please take this last chance to review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the cover letter that accompanies your revised manuscript.

In addition to these revisions, you may need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests shortly. If you do not receive a separate email within a few days, please assume that checks have been completed, and no additional changes are required.

We expect to receive your revised manuscript within two weeks.

To submit your revision, please go to https://www.editorialmanager.com/pbiology/ and log in as an Author. Click the link labelled 'Submissions Needing Revision' to find your submission record. Your revised submission must include the following:

- a cover letter that should detail your responses to any editorial requests, if applicable, and whether changes have been made to the reference list

- a Response to Reviewers file that provides a detailed response to the reviewers' comments (if applicable, if not applicable please do not delete your existing 'Response to Reviewers' file.)

- a track-changes file indicating any changes that you have made to the manuscript.

NOTE: If Supporting Information files are included with your article, note that these are not copyedited and will be published as they are submitted. Please ensure that these files are legible and of high quality (at least 300 dpi) in an easily accessible file format. For this reason, please be aware that any references listed in an SI file will not be indexed. For more information, see our Supporting Information guidelines:

https://journals.plos.org/plosbiology/s/supporting-information

*Published Peer Review History*

Please note that you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. Please see here for more details:

https://plos.org/published-peer-review-history/

*Press*

Should you, your institution's press office or the journal office choose to press release your paper, please ensure you have opted out of Early Article Posting on the submission form. We ask that you notify us as soon as possible if you or your institution is planning to press release the article.

*Protocols deposition*

Please do not hesitate to contact me should you have any questions.

Sincerely,

Melissa

Melissa Vazquez Hernandez, Ph.D.

Associate Editor

mvazquezhernandez@plos.org

PLOS Biology

------------------------------------------------------------------------

REVIEWERS' COMMENTS:

------------------------------------------------------------------------

Reviewer #1: Apologies for taking a couple extra days on the review. After looking at the resubmitted materials, the authors have done a nice job of shifting the focus further towards inter-lab reproducibility, including addressing that statistically. They have also clarified my earlier points of confusion.

The authors have asked "For your convenience below is a figure incorporating the number of amplification cycles to clarify that these reflect the preparation of the different sample types (30 PCR cycles for Media and 22 cycles for Root to achieve similar DNA yield). Let us know if you see a value in including it in the manuscript. We believe that a simple statement in M&M should be sufficient:"

I agree with the authors that the statement they go on to propose is sufficient. I think the paper can be accepted.

------------------------------------------------------------------------

Reviewer #2 (Mengcen Wang): The authors have addressed my concerns, and I therefore recommend its publication.

PLoS Biol. doi: 10.1371/journal.pbio.3003358.r005

Decision Letter 3

Melissa Vazquez Hernandez

10 Aug 2025

Dear Trent,

Thank you for the submission of your revised Meta-Research Article "Breaking the reproducibility barrier with standardized protocols for plant-microbiome research" for publication in PLOS Biology. On behalf of my colleagues and the Academic Editor, Cara Haney, I am pleased to say that we can in principle accept your manuscript for publication, provided you address any remaining formatting and reporting issues. These will be detailed in an email you should receive within 2-3 business days from our colleagues in the journal operations team; no action is required from you until then. Please note that we will not be able to formally accept your manuscript and schedule it for publication until you have completed any requested changes.

Please take a minute to log into Editorial Manager at http://www.editorialmanager.com/pbiology/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process.

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with biologypress@plos.org. If you have previously opted in to the early version process, we ask that you notify us immediately of any press plans so that we may opt out on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

Thank you again for choosing PLOS Biology for publication and supporting Open Access publishing. We look forward to publishing your study.

Sincerely,

Melissa

Melissa Vazquez Hernandez, Ph.D., Ph.D.

Associate Editor

PLOS Biology

mvazquezhernandez@plos.org

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Plant biomass data for laboratories A–E.

(TIFF)

pbio.3003358.s001.tiff^{(924.7KB, tiff)}

S2 Fig. Plant growth conditions in labs A–E.

(TIFF)

pbio.3003358.s002.tiff^{(5.5MB, tiff)}

S3 Fig. Representative images of root systems.

(TIFF)

pbio.3003358.s003.tiff^{(8.6MB, tiff)}

S4 Fig. Microbiome composition.

(TIFF)

pbio.3003358.s004.tiff^{(1.5MB, tiff)}

S5 Fig. Untargeted metabolomics on rhizosphere metabolites.

(TIFF)

pbio.3003358.s005.tiff^{(1.2MB, tiff)}

S6 Fig. Statistical analysis of 16S rRNA sequencing data.

(TIFF)

pbio.3003358.s006.tiff^{(1.8MB, tiff)}

S7 Fig. Statistical analysis of 60 identified rhizosphere metabolites.

(TIFF)

pbio.3003358.s007.tiff^{(2MB, tiff)}

S8 Fig. Comparative genomics of bacterial isolates.

(TIFF)

pbio.3003358.s008.tiff^{(2.7MB, tiff)}

S9 Fig. Fluorescent microscopy in EcoFAB 2.0 conducted on plants grown in the lab A.

(TIFF)

pbio.3003358.s009.tiff^{(8.2MB, tiff)}

S10 Fig. Assessment of growth and biofilm formation for bacterial isolates in vitro by the lab A.

(TIFF)

pbio.3003358.s010.tiff^{(933.3KB, tiff)}

S11 Fig. Motility assays in the lab A.

(TIFF)

pbio.3003358.s011.tiff^{(5MB, tiff)}

S1 Table. Chamber settings and models and experiment timing.

(XLSX)

pbio.3003358.s012.xlsx^{(71.1KB, xlsx)}

S2 Table. SynCom members overview, genome information, and their OD₆₀₀ to CFU conversion ratios.

(XLSX)

pbio.3003358.s013.xlsx^{(15.7KB, xlsx)}

S3 Table. 16S rRNA sequencing read counts in media and root samples.

Samples labeled “none” are Axenic plant controls, “Gly” are glycerol stocks of inoculum.

(XLSX)

pbio.3003358.s014.xlsx^{(51.4KB, xlsx)}

S4 Table. BLAST search results for unknown reads in selected media samples.

(XLSX)

pbio.3003358.s015.xlsx^{(13.1KB, xlsx)}

S5 Table. LC–MS/MS parameters.

(XLSX)

pbio.3003358.s016.xlsx^{(11.8KB, xlsx)}

S6 Table. Metabolite identification and intensity.

(XLSX)

pbio.3003358.s017.xlsx^{(175.9KB, xlsx)}

S7 Table. ANOVA results for 16S rRNA sequencing data and metabolite intensity.

(XLSX)

pbio.3003358.s018.xlsx^{(53.3KB, xlsx)}

S8 Table. Pairwise results for metabolites.

(XLSX)

pbio.3003358.s019.xlsx^{(213.9KB, xlsx)}

S9 Table. Comparison of KEGG modules between SynCom16 vs. Paraburkholderia sp. OAS925.

(XLSX)

pbio.3003358.s020.xlsx^{(35.6KB, xlsx)}

S1 Video. Fluorescent microscopy at 1 DAI.

Z-Stack shows RFP-Paraburkholderia motility and colonization on B. distachyon roots at 1 DAI in EcoFAB 2.0.

(MP4)

Download video file^{(8.8MB, mp4)}

S2 Video. Fluorescent microscopy at 3 DAI.

Root colonization by RFP-Paraburkholderia at 3 DAI in EcoFAB 2.0. Merged bright-field and TxRed channels followed by TxRed footage.

(MP4)

Download video file^{(11.5MB, mp4)}

S1 File. MZMine 3 settings.

(ZIP)

pbio.3003358.s023.zip^{(3.8KB, zip)}

Attachment

Submitted filename: Responses.pdf

pbio.3003358.s026.pdf^{(9.7MB, pdf)}

Data Availability Statement

[pbio.3003358.ref001] 1.Zengler K, Hofmockel K, Baliga NS, Behie SW, Bernstein HC, Brown JB, et al. EcoFABs: advancing microbiome science through standardized fabricated ecosystems. Nat Methods. 2019;16(7):567–71. doi: 10.1038/s41592-019-0465-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref002] 2.O’Toole GA. We have a community problem. J Bacteriol. 2024;206(4):e0007324. doi: 10.1128/jb.00073-24 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref003] 3.Vorholt JA, Vogel C, Carlström CI, Müller DB. Establishing causality: opportunities of synthetic communities for plant microbiome research. Cell Host Microbe. 2017;22(2):142–55. doi: 10.1016/j.chom.2017.07.004 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref004] 4.Marín O, González B, Poupin MJ. From microbial dynamics to functionality in the rhizosphere: a systematic review of the opportunities with synthetic microbial communities. Front Plant Sci. 2021;12:650609. doi: 10.3389/fpls.2021.650609 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref005] 5.Salas-González I, Reyt G, Flis P, Custódio V, Gopaulchan D, Bakhoum N, et al. Coordination between microbiota and root endodermis supports plant mineral nutrient homeostasis. Science. 2021;371(6525):eabd0695. doi: 10.1126/science.abd0695 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref006] 6.Finkel OM, Salas-González I, Castrillo G, Conway JM, Law TF, Teixeira PJPL, et al. A single bacterial genus maintains root growth in a complex microbiome. Nature. 2020;587(7832):103–8. doi: 10.1038/s41586-020-2778-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref007] 7.Vandenkoornhuyse P, Quaiser A, Duhamel M, Le Van A, Dufresne A. The importance of the microbiome of the plant holobiont. New Phytol. 2015;206(4):1196–206. doi: 10.1111/nph.13312 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref008] 8.Northen TR, Kleiner M, Torres M, Kovács ÁT, Nicolaisen MH, Krzyżanowska DM. Community standards and future opportunities for synthetic communities in plant-microbiota research. Nat Microbiol. 2024;9(11):2774–84. doi: 10.1038/s41564-024-01833-4 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref009] 9.Coker J, Zhalnina K, Marotz C, Thiruppathy D, Tjuanta M, D’Elia G, et al. A reproducible and tunable synthetic soil microbial community provides new insights into microbial ecology. mSystems. 2022;7(6):e0095122. doi: 10.1128/msystems.00951-22 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref010] 10.Sasse J, Kant J, Cole BJ, Klein AP, Arsova B, Schlaepfer P, et al. Multilab EcoFAB study shows highly reproducible physiology and depletion of soil metabolites by a model grass. New Phytol. 2019;222(2):1149–60. doi: 10.1111/nph.15662 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref011] 11.Lin H-H, Torres M, Adams CA, Andeer PF, Owens TK, Zhalnina K, et al. Impact of inoculation practices on microbiota assembly and community stability in a fabricated ecosystem. Phytobiomes J. 2024;8(2):155–67. doi: 10.1094/pbiomes-06-23-0050-r [DOI] [Google Scholar]

[pbio.3003358.ref012] 12.Novak V, Andeer PF, Bowen BP, Ding Y, Zhalnina K, Hofmockel KS, et al. Reproducible growth of Brachypodium in EcoFAB 2.0 reveals that nitrogen form and starvation modulate root exudation. Sci Adv. 2024;10(1):eadg7888. doi: 10.1126/sciadv.adg7888 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref013] 13.McGrath TF, Haughey SA, Islam M, Elliott CT, Collaborators. The potential of handheld near infrared spectroscopy to detect food adulteration: results of a global, multi-instrument inter-laboratory study. Food Chem. 2021;353:128718. doi: 10.1016/j.foodchem.2020.128718 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref014] 14.Thompson M, Ellison SLR, Wood R. The International Harmonized Protocol for the proficiency testing of analytical chemistry laboratories (IUPAC Technical Report). Pure Appl Chem. 2006;78(1):145–96. doi: 10.1351/pac200678010145 [DOI] [Google Scholar]

[pbio.3003358.ref015] 15.Novak V, King E, Andeer P, Vogel J, t Northen T. Use of EcoFAB 2.0 for reproducible plant-microbe interaction experiments v1. 2024. Available from: doi: 10.17504/protocols.io.kxygxyydkl8j/v1 [DOI] [Google Scholar]

[pbio.3003358.ref016] 16.Bowen B. SynCom17_EcoFab. Hugging Face. 2025. doi: 10.57967/hf/5885 [DOI] [Google Scholar]

[pbio.3003358.ref017] 17.Zhalnina K, Louie KB, Hao Z, Mansoori N, da Rocha UN, Shi S, et al. Dynamic root exudate chemistry and microbial substrate preferences drive patterns in rhizosphere microbial community assembly. Nat Microbiol. 2018;3(4):470–80. doi: 10.1038/s41564-018-0129-3 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref018] 18.Korenblum E, Dong Y, Szymanski J, Panda S, Jozwiak A, Massalha H, et al. Rhizosphere microbiome mediates systemic root metabolite exudation by root-to-root signaling. Proc Natl Acad Sci U S A. 2020;117(7):3874–83. doi: 10.1073/pnas.1912130117 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref019] 19.Allison SD, Vitousek PM. Responses of extracellular enzymes to simple and complex nutrient inputs. Soil Biol Biochemi. 2005;37(5):937–44. doi: 10.1016/j.soilbio.2004.09.014 [DOI] [Google Scholar]

[pbio.3003358.ref020] 20.de Raad M, Li YV, Kuehl JV, Andeer PF, Kosina SM, Hendrickson A, et al. A defined medium for cultivation and exometabolite profiling of soil bacteria. Front Microbiol. 2022;13:855331. doi: 10.3389/fmicb.2022.855331 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref021] 21.Schloss PD. Identifying and overcoming threats to reproducibility, replicability, robustness, and generalizability in microbiome research. mBio. 2018;9(3):e00525–18. doi: 10.1128/mBio.00525-18 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref022] 22.Meiser J, Weindl D, Hiller K. Complexity of dopamine metabolism. Cell Commun Signal. 2013;11(1):34. doi: 10.1186/1478-811X-11-34 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref023] 23.Umek N, Geršak B, Vintar N, Šoštarič M, Mavri J. Dopamine autoxidation is controlled by acidic pH. Front Mol Neurosci. 2018;11:467. doi: 10.3389/fnmol.2018.00467 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref024] 24.Nuccio EE, Starr E, Karaoz U, Brodie EL, Zhou J, Tringe SG, et al. Niche differentiation is spatially and temporally regulated in the rhizosphere. ISME J. 2020;14(4):999–1014. doi: 10.1038/s41396-019-0582-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref025] 25.Fierer N. Embracing the unknown: disentangling the complexities of the soil microbiome. Nat Rev Microbiol. 2017;15(10):579–90. doi: 10.1038/nrmicro.2017.87 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref026] 26.Debray R, Herbert RA, Jaffe AL, Crits-Christoph A, Power ME, Koskella B. Priority effects in microbiome assembly. Nat Rev Microbiol. 2022;20(2):109–21. doi: 10.1038/s41579-021-00604-w [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref027] 27.Carlström CI, Field CM, Bortfeld-Miller M, Müller B, Sunagawa S, Vorholt JA. Synthetic microbiota reveal priority effects and keystone strains in the Arabidopsis phyllosphere. Nat Ecol Evol. 2019;3(10):1445–54. doi: 10.1038/s41559-019-0994-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref028] 28.Kuzyakov Y, Razavi BS. Rhizosphere size and shape: temporal dynamics and spatial stationarity. Soil Biol Biochem. 2019;135:343–60. doi: 10.1016/j.soilbio.2019.05.011 [DOI] [Google Scholar]

[pbio.3003358.ref029] 29.Lu P, Ma D, Chen Y, Guo Y, Chen G-Q, Deng H, et al. L-glutamine provides acid resistance for Escherichia coli through enzymatic release of ammonia. Cell Res. 2013;23(5):635–44. doi: 10.1038/cr.2013.13 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref030] 30.Minamino T, Imae Y, Oosawa F, Kobayashi Y, Oosawa K. Effect of intracellular pH on rotational speed of bacterial flagellar motors. J Bacteriol. 2003;185(4):1190–4. doi: 10.1128/JB.185.4.1190-1194.2003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref031] 31.Cole BJ, Feltcher ME, Waters RJ, Wetmore KM, Mucyn TS, Ryan EM, et al. Genome-wide identification of bacterial plant colonization genes. PLoS Biol. 2017;15(9):e2002860. doi: 10.1371/journal.pbio.2002860 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref032] 32.Hafner K, Welt B, Pelletier W. Dry ice sublimation performance as affected by binding agent, density, and age. Engineering. 2023. [Google Scholar]

[pbio.3003358.ref033] 33.Teytelman L, Stoliartchouk A, Kindler L, Hurwitz BL. Protocols.io: virtual communities for protocol development and discussion. PLoS Biol. 2016;14(8):e1002538. doi: 10.1371/journal.pbio.1002538 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref034] 34.Eloe-Fadrosh EA, Ahmed F, Babinski M, Baumes J, Borkum M, Bramer L, et al. The National Microbiome Data Collaborative Data Portal: an integrated multi-omics microbiome data resource. NAT. 2022. doi: 10.1093/nar/gkab990 [DOI] [Google Scholar]

[pbio.3003358.ref035] 35.Sinha R, Abu-Ali G, Vogtmann E, Fodor AA, Ren B, Amir A, et al. Assessment of variation in microbial community amplicon sequencing by the Microbiome Quality Control (MBQC) project consortium. Nat Biotechnol. 2017;35(11):1077–86. doi: 10.1038/nbt.3981 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref036] 36.Castrillo G, Teixeira PJPL, Paredes SH, Law TF, de Lorenzo L, Feltcher ME, et al. Root microbiota drive direct integration of phosphate stress and immunity. Nature. 2017;543(7646):513–8. doi: 10.1038/nature21417 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref037] 37.Durán P, Thiergart T, Garrido-Oter R, Agler M, Kemen E, Schulze-Lefert P, et al. Microbial interkingdom interactions in roots promote Arabidopsis survival. Cell. 2018;175(4):973-983.e14. doi: 10.1016/j.cell.2018.10.020 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref038] 38.Sordo Z, Andeer P, Sethian J, Northen T, Ushizima D. RhizoNet segments plant roots to assess biomass and growth for enabling self-driving labs. Sci Rep. 2024;14(1):12907. doi: 10.1038/s41598-024-63497-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref039] 39.Lobet G, Pagès L, Draye X. A novel image-analysis toolbox enabling quantitative analysis of root system architecture. Plant Physiol. 2011;157(1):29–39. doi: 10.1104/pp.111.179895 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref040] 40.Schmid R, Heuckeroth S, Korf A, Smirnov A, Myers O, Dyrlund TS, et al. Integrative analysis of multimodal mass spectrometry data in MZmine 3. Nat Biotechnol. 2023;41(4):447–9. doi: 10.1038/s41587-023-01690-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref041] 41.Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, et al. Sharing and community curation of mass spectrometry data with Global Natural Products Social Molecular Networking. Nat Biotechnol. 2016;34(8):828–37. doi: 10.1038/nbt.3597 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref042] 42.Bowen BP, Northen TR. Dealing with the unknown: metabolomics and metabolite atlases. J Am Soc Mass Spectrom. 2010;21(9):1471–6. doi: 10.1016/j.jasms.2010.04.003 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref043] 43.Shaffer JP, Nothias L-F, Thompson LR, Sanders JG, Salido RA, Couvillion SP, et al. Standardized multi-omics of Earth’s microbiomes reveals microbial and metabolite diversity. Nat Microbiol. 2022;7(12):2128–50. doi: 10.1038/s41564-022-01266-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref044] 44.Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460–1. doi: 10.1093/bioinformatics/btq461 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref045] 45.Haney EF, Trimble MJ, Hancock REW. Microtiter plate assays to assess antibiofilm activity against bacteria. Nat Protoc. 2021;16(5):2615–32. doi: 10.1038/s41596-021-00515-3 [DOI] [PubMed] [Google Scholar]

[pbio.3003358.ref046] 46.Mukherjee S, Stamatis D, Li CT, Ovchinnikova G, Bertsch J, Sundaramurthi JC, et al. Twenty-five years of Genomes OnLine Database (GOLD): data updates and new features in v.9. Nucleic Acids Res. 2023;51(D1):D957–63. doi: 10.1093/nar/gkac974 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref047] 47.Chen I-MA, Chu K, Palaniappan K, Ratner A, Huang J, Huntemann M, et al. The IMG/M data management and analysis system v.7: content updates and new features. Nucleic Acids Res. 2023;51(D1):D723–32. doi: 10.1093/nar/gkac976 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref048] 48.Chen I-MA, Chu K, Palaniappan K, Ratner A, Huang J, Huntemann M, et al. The IMG/M data management and analysis system v.6.0: new tools and advanced capabilities. Nucleic Acids Res. 2021;49(D1):D751–63. doi: 10.1093/nar/gkaa939 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref049] 49.Parks DH, Chuvochina M, Rinke C, Mussig AJ, Chaumeil P-A, Hugenholtz P. GTDB: an ongoing census of bacterial and archaeal diversity through a phylogenetically consistent, rank normalized and complete genome-based taxonomy. Nucleic Acids Res. 2022;50: D785–D794. doi: 10.1093/nar/gkab776 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref050] 50.Price MN, Dehal PS, Arkin AP. FastTree 2—approximately maximum-likelihood trees for large alignments. PLoS One. 2010;5(3):e9490. doi: 10.1371/journal.pone.0009490 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref051] 51.Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 2021;49(W1):W293–6. doi: 10.1093/nar/gkab301 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref052] 52.Pearson AN, Thompson MG, Kirkpatrick LD, Ho C, Vuu KM, Waldburger LM, et al. The pGinger family of expression plasmids. Microbiol Spectr. 2023;11(3):e0037323. doi: 10.1128/spectrum.00373-23 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pbio.3003358.ref053] 53.Warnes GR, Bolker B, Bonebakker L, Gentleman R, Huber W, Liaw A, et al. gplots: Various R Programming Tools for Plotting Data. R package version 3.1.3.1. 2024 [cited 31 Mar 2024]. Available from: https://CRAN.R-project.org/package=gplots

[pbio.3003358.ref054] 54.Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community Ecology Package. 2020. Available from: http://CRAN.R-project.org/package=vegan

PERMALINK

Breaking the reproducibility barrier with standardized protocols for plant–microbiome research

Vlastimil Novak

Peter F Andeer

Eoghan King

Jacob Calabria

Connor Fitzpatrick

Jana M Kelm

Kathrin Wippel

Suzanne M Kosina

Benjamin P Bowen

Chris Daum

Matthew Zane

Archana Yadav

Mingfei Chen

Dor Russ

Catharine A Adams

Trenton K Owens

Bradie Lee

Yezhang Ding

Zineb Sordo

Romy Chakraborty

Simon Roux

Adam M Deutschbauer

Daniela Ushizima

Karsten Zengler

Borjana Arsova

Jeffery L Dangl

Paul Schulze-Lefert

Michelle Watt

John P Vogel

Trent R Northen

Roles

Abstract

1. Introduction

2. Results

2.1. Study design and logistics

Fig 1. Overview of the interlaboratory comparison study.

2.2. Protocols resulted in reproducibly sterile conditions

Fig 2. Plant phenomics and EcoFAB 2.0 device sterility.

2.3. Reproducible plant growth

2.4. Reproducible microbiome assembly

Fig 3. Root microbiome.

2.5. Reproducible rhizosphere metabolome

Fig 4. Targeted metabolomic analysis of rhizosphere.

2.6. Statistical evaluation of variation between labs

2.7. Colonization by Paraburkholderia

3. Discussion

Fig 5. Framework for conducting reproducible multi-lab microbiome studies.

4. Materials and methods

4.1. Preparation of synthetic bacterial communities

4.2. Experimental setup and plant growth conditions

4.3. Root phenotyping

4.4. Samples collection and shipment

4.5. Analysis of rhizosphere metabolites

4.6. Microbiome analysis by amplicon sequencing

4.7. Growth and biofilm formation assays for bacterial isolates

4.8. Genomic analysis of bacterial isolates

4.9. Fluorescent microscopy

4.10. Motility assays

4.11. Data management, statistical analyses, software, and data visualization

Supporting information

Acknowledgments

Abbreviations

:

Data Availability

Funding Statement

References

Decision Letter 0

Melissa Vazquez Hernandez

Roles

Decision Letter 1

Melissa Vazquez Hernandez

Roles

Author response to Decision Letter 2

Decision Letter 2

Melissa Vazquez Hernandez

Roles

Decision Letter 3

Melissa Vazquez Hernandez