ABLeS Participant: School of Agriculture, Food and Wine, The University of Adelaide

Project title

Global Functional Shifts in Wheat-Field Soils Under Drought vs Irrigation

Collaborators and funding

School of Agriculture, Food and Wine, The University of Adelaide, Urrbrae, SA, Australia

Contact(s)

• Jiayu Li, School of Agriculture, Food and Wine, The University of Adelaide, Jiayu.li02@adelaide.edu.au

Project description and aims

1) Project summary Goal. Quantify how drought (rainfed) vs irrigation shapes the functional capacity of wheat-field soil microbiomes at global scale, then zoom in on a deep-sequenced Australian transect to resolve fine-grained pathways, genomes (MAGs), and ecological indicators. Datasets. Global set: ~300 bulk-soil metagenomes (each ~20 Gb) from wheat fields worldwide (rainfed vs irrigated), plus amplicons (16S/ITS) where available. Australian transect: 68 bulk-soil samples spanning west→east; each ~40 Gb metagenome (+ matched amplicons). Overall approach. Harmonise metadata and processing; perform function- and genome-resolved metagenomics; build drought/irrigation classifiers and indicator gene panels; validate patterns with the Australian transect; release an open resource (MAGs, gene catalogues, reproducible workflows). 2) Aims & testable hypotheses Aim 1 Global functional contrasts. Quantify differences in metabolic potential between rainfed vs irrigated wheat soils. H1: Droughted (rainfed) sites show enrichment of osmoprotection (e.g., trehalose/betaine), EPS/aggregate formation, ROS mitigation, and water-use efficiency-linked nitrogen cycling routes; irrigated sites show higher denitrification potential and copiotrophic traits. Aim 2 Genome-resolved ecology. Recover and compare MAGs and strain populations associated with drought vs irrigation. H2: Distinct DefenceBiome-like consortia (e.g., Actinobacteria, Streptomyces, Microbacteriaceae) increase in drought and encode exudate-responsive transporters and stress regulons. Aim 3 Indicator panels & predictive models. Develop gene/pathway and MAG indicator sets that predict water regime and agronomic context across continents. H3: A small panel of functions (≤50 KOs/CAZymes) plus a handful of sentinel MAGs will classify water regime with high accuracy (AUC > 0.85) after controlling for soil/climate covariates. Aim 4 Australian transect validation & mechanistic resolution. Use the 68-site transect to (i) validate global signatures; (ii) resolve fine-grained pathway variants, strain microdiversity, and ecological thresholds along rainfall/irrigation gradients. 3) Significance & impact

• Agronomic relevance: actionable microbiome indicators for drought-smart management, informing irrigation scheduling, residue/cover practices, and microbial amendments.
• Scientific advance: links host water regime → exudates → microbial functions/MAGs, uniting community and genome-resolved views at continental scale.
• Community resource: openly released MAG catalogue, non-redundant gene set, and reproducible pipelines, enabling re-use in wheat microbiome, soil health, and climate-adaptation studies. 4) Planned analyses
• 4.1 Data audit & harmonisation
  • Inventory all cohorts; standardise metadata schema (soil chem/texture, climate, management, cultivar, water regime, geography).
  • Define inclusion criteria; handle licence/consent; map to MIxS-Soil fields. Deliverable: harmonised metadata table; PRISMA-style flow diagram.
• 4.2 Read processing & profiling
  • QC: FastQC/MultiQC → adapter/quality trimming → host (wheat) removal.
  • Taxon/function profiling (reads): Kraken2/Bracken (or centrifuge) + HUMAnN/eggNOG-mapper for KO/EC pathways
  • ShortBRED for targeted marker sets (e.g., osmolyte genes). Deliverable: per-sample taxon and functional tables + QC reports.
• 4.3 Assembly & gene catalogues
  • Strategy chosen per cohort (single/co-assembly by eco-region); MEGAHIT/metaSPAdes with k-mer sweeps; protein prediction (Prodigal).
  • Non-redundant gene catalogue via MMseqs2; function annotation (KEGG/KO, EC, COG, CAZy, antiSMASH/BGCs); resistome where relevant. Deliverable: gene catalogue (FASTA + annotation TSV), abundance matrices.
• 4.4 MAG reconstruction & curation
  • Binning with MetaBAT2, CONCOCT, VAMB; refinement (DASTool); quality via CheckM2; dereplication (dRep); taxonomy (GTDB-Tk).
  • Abundance/coverage via coverM; SNP/strain tracking (inStrain). Deliverable: dereplicated MAG set (≥50% comp, ≤10% contam; high-quality subset ≥90/≤5), per-MAG metadata.
• 4.5 Pathway & network ecology
  • Targeted pathways: osmolytes (proline/betaine/trehalose), EPS, antioxidant systems, nitrogen (nitrification, denitrification, DNRA), sulfur shunts, phosphorus solubilisation, ACC deaminase, transporters for key exudates.
  • Co-occurrence & guild inference (FastSpar/SpiecEasi) → stability/robustness tests. Deliverable: pathway differential analyses, guild maps, effect sizes.
• 4.6 Statistics & causal controls
  • Compositional methods (ALDEx2/ANCOM-BC) with covariate control (soil pH, texture, organic C, MAP/MAT, continent, cultivar) via mixed-effects models.
  • Counterfactual checks (matching/propensity) to mitigate irrigation vs region confounding. Deliverable: adjusted global contrasts with uncertainty quantification.
• 4.7 Predictive modelling & indicators
  • ML (elastic net / random forest / XGBoost) to predict water regime and agronomic outcomes from functions + MAGs; nested CV and external validation (Australian transect).
  • Derive minimal indicator panels for field diagnostics. Deliverable: classifiers (AUC/PR curves), top features, portable panels.
• 4.8 Australian transect deep-dive
  • High-resolution assembly, strain-level microdiversity, fine-scale pathway variants; change-point analyses along rainfall/irrigation; link to soil properties. Deliverable: Australia-focused paper validating global signals.

How is ABLeS supporting this work?

This work is supported through the Production Bioinformatics scheme provided by ABLeS. The support includes storage and compute allocation.

Expected outputs enabled by participation in ABLeS

• D1. Global non-redundant gene catalogue for wheat-field soils (KO/EC/CAZy/BGC).
• D2. Curated MAG set with habitat preferences and trait annotations.
• D3. Indicator panels (genes/MAGs) and validated predictive models of water regime.
• D4. Reproducible Nextflow workflows and environment recipes (Apptainer).
• D5. High impact papers: (i) global functional contrasts; (ii) genome-resolved validation on the Australian transect; (iii) workflow/methods.
• D6. FAIR-compliant data/metadata releases with DOIs.

These details have been provided by project members at project initiation. For more information on the project, please consult the contact(s) or project links above.