Data from: Breeding of microbiomes conferring salt tolerance to plants

Guilherme Pereira, Caio 1 ; Edwards, Joseph2 ; Khasanova, Albina 1 ; Carlson, Alexis1 ; Brisson, Vanessa3 ; Schaefer, Estelle1 ; Glavina del Rio, Tijana3 ; Tringe, Susannah3 ; Vogel, John 3 ; Des Marais, David 4 ; Juenger, Thomas 1 ; Mueller, Ulrich1

Research facility: The University of Texas at Austin

Published Jan 28, 2026 on Dryad. https://doi.org/10.5061/dryad.h44j0zpwg

Data files

Jan 28, 2026 version files 6.74 MB

data_Bd_G9.rds

6.61 MB
G9_Metadata.csv

20.06 KB
Ionomics_Brachy.csv

99.48 KB
README.md

6.33 KB

Abstract

Microbiome breeding by host-mediated selection is a technique to artificially select for microbiomes conferring benefits to plants. Here, we describe leaf ionomics, microbial community composition, and network analyses of a microbiome-breeding experiment to generate microbiomes conferring salt tolerance to Brachypodium distachyon, a model for cereal crops. Plants receiving microbiomes selected to confer tolerance to either sodium- or aluminium-stress produced 69-198% higher total seed weight than plants receiving control microbiomes. Sodium-selected microbiomes reduced leaf-sodium concentration by 50%, whereas aluminium-selected microbiomes had no effect on leaf-tissue nutrient concentration, suggesting different mechanisms underlying microbiome-mediated salt tolerance. By testing these selected microbiomes in a cross-fostering experiment, we show that artificially-selected microbiomes attain (a) ecological robustness contributing to transplantability (inheritance) of microbiome-encoded effects between plants; and (b) network features identifying key bacteria promoting stress tolerance. Combined, these findings elucidate critical mechanisms underlying host-mediated selection as a tool to breed beneficial microbiomes in an agricultural context.

https://doi.org/10.5061/dryad.h44j0zpwg

Description of the data and file structure

Data and scripts used in the analyses presented in the manuscript "Breeding of microbiomes conferring salt tolerance to plants"

* Details regarding the collection of data (as well as subsequent analyses) are provided in detail in the published manuscript (https://doi.org/10.1186/s40168-025-02261-0) and in Mueller et al 2021 ("Artificial selection on microbiomes to breed microbiomes that confer salt tolerance to plants", mSystems Vol. 6, No. 6; doi.org/10.1128/mSystems.01125-21). The Supplementary Methods of Mueller et al 2021, in particular file msystems.01125-21-t0001.pdf, has extensive details on the experimental design and protocols. For any questions related to data and/or analyses, please contact Caio Guilherme Pereira (caiogp@utexas.edu / www.cguilhermepereira.com/).

Datasets are organized as following:

data_Bd_G9.rds (Data) - Dataset with the results of the 16S sequencing used throughout the manuscript.

The .rds file contains three datasets: 1.) a "cts" dataset with the read count for each ASV; 2.) a "tax" dataset with taxonomic information (from kingdom to genera), along with the 16S rRNA amplicon sequence for each ASV; and 3.) an "info" dataset with experimental design information such as generation, selection line, habitat, etc. (the variables described here match those of the G9_Metadata.csv, described below).

G9_Metadata.csv (Data) - Dataset with detailed description of samples from the cross-fostering experiment (Gen 9). These include the following information:

Plant ID (numerical, with a unique identifier for each plant);
Selection History (categorical; 'SOD' identifies sodium-exposed and 'ALU' identifies aluminum-exposed plants during the selection process);
Salt Stress (categorical; 'SOD' identifies sodium-stressed and 'ALU' identifies aluminum-stressed plants during the cross-fostering experiment);
Microbiome Treatment (categorical; 'Pp' identifies plants that received a selected microbiome inoculum; 'PpFilt' identifies plants that received a double-filtered inoculum with no live cells - i.e., the solute control; 'Np' identifies plants that received a inoculum extracted from plant-free pots - i.e., the fallow-soil control; and 'Null' identifies plants that did not receive a microbial inoculum);
Selection Line (numerical, ranging from 1 to 5 and identifying the 5 selection lines established throughout the selection process);
Rack Number (numerical, ranging from 1-8 and identifying the rack number in which the plant was placed during Gen 9; this is a co-variate used in the linear mixed-effect modelling to account for possible spatial biases);
Rack Position (numerical, identifying the position of the rack within the growth chamber during Gen 9; this is also a co-variate used in the linear mixed-effect modelling to account for possible spatial biases);
Seed Weight (numerical, identifying the initial seed weight (in mg) for each individual plant in Gen 9; this is another co-variate used in the linear mixed-effect modelling to account for possible initial size effects).

Ionomics_Brachy.csv (Data) - Dataset with the leaf-tissue nutrient concentration of samples from the cross-fostering experiment (Gen 9). It also includes the following information regarding the ICP-MS (Inductively Coupled Plasma Mass Spectrometry) runs:

sample (numerical, with a unique identifier for each plant);
numinSet (numerical; describes the tube number, within the set of 288 samples in a WR.run, in which the sample was contained);
fullnum (numerical; describes the order in which the sample was analyzed throughout the entire experiment);
WR.run (numerical; describes the ‘weighing robot’ batch of 288 samples that was analyzed in each run);
run.date (categorical; describes the exact time and date in which that sample was analyzed);
SampleWeight (numerical; describes the amount of leaf tissue, expressed in mg, that was used for analysis);
The leaf-tissue concentration (expressed in μg/g) of each element analyzed is presented from columns 'G' to 'Y', each representing a particular element and named with the combination of the element symbol and its atomic number (e.g., sodium is Na23). The columns are ordered from lowest to highest atomic number.

In addition to these datasets, the following scripts are being made public in a Zenodo publication (connected, but separate from this Dryad data publication);

Plant_Fitness_(Seed_Weigh).R (Script) - R script for the analyses of plant fitness (presented as total seed weight; mg) presented in Figs 2b and 3b.

Leaf_Ionomics_(Leaf_Na_and_Al).R (Script) - R script for the analyses of leaf-sodium and -aluminium concentrations presented in Fig. 3.

Microbial_Comm_Analyses_(G9).R (Script) - R script for the analyses of microbial community composition presented in Figs 4, S1, and S2.

Network_Run_(sparCC).R (Script) - R script for the estimation of microbial co-occurrence networks presented in Figs 5 and S3, Tables 1 and S1.

Network_Analyses_(G9).R (Script) - R script for the analyses of microbial co-occurrence networks presented in Figs 5 and S3, Tables 1 and S1.

Leaf_Ionomics_(Macronutrients).R (Script) - R script for the analyses of leaf-macronutrient concentrations presented in Fig. S4.

Leaf_Ionomics_(Micronutrients).R (Script) - R script for the analyses of leaf-micronutrient concentrations presented in Fig. S5.

Leaf_Ionomics_(Trace_Elements).R (Script) - R script for the analyses of leaf trace-element concentrations presented in Fig. S6.

Code/software

All packages required to perform the analyses described in the manuscript are provided within their respective scripts. Each script contains all information required for their unique analysis (and resulting figure) in a step-by-step fashion, with comprehensive notes throughout the code to help the reader/user.