Plant abiotic stress methylation pack (example)¶

Instance of the methylation application-pack pattern (trait application). A binary Control vs Drought Arabidopsis thaliana leaf WGBS methylation study on the existing methylation control plane. It selects assay procedure plant_wgbs_gene_fc (linear WGBS, gene FeatureCuts, plant lifecycle without blood deconvolution), the samd_research profile, and the plant_tissue analyte profile. The pack is config + cohorts + partitions + a trait overlay plus a non-human site reference — no new actions or aligners.

This instance extends the application-pack pattern with plant platform unblockers: the plant_tissue analyte (plant-safe QC that treats CG/CHG/CHH as real biology, no cfDNA fragmentomics), a linear TAIR10 site reference, a lifecycle program without blood cell deconvolution, and the plant-stress-core enrichment preset.

See the operator guide: Usage ch.23 Plant abiotic stress pack, the analyte defaults in ANALYTE_PROFILES, and the plant epigenomics background in Plant Research.

Files¶

File	Role
`project_Control_vs_Drought.json`	Default (Arabidopsis): Control vs Drought, chroms 1–5
`context_plant_abiotic_stress.json`	Arabidopsis overlay: `pipelineProcedure=plant_wgbs_gene_fc`, `plant_traits` prior + `plant-stress-core`, `string_species=3702`
`project_Control_vs_Drought_{soybean,maize,wheat}.json`	Crop manifests (same binary design; crop chromosome lists)
`context_{soybean,maize,wheat}_drought.json`	Crop overlays (`string_species` 3847 / 4577 / 4565)
`data/*.csv`	Shared cohort CSV stubs (replace with real, plant-disjoint sample IDs)
`data/arabidopsis_drought_gene_traits.tsv`	Demo offline gene↔trait prior (`enrich_source=plant_traits`)

Provision a plant genome (once per cluster)¶

Plants align linear only (no HPRC pangenome). Pick one site recipe:

Species	Download script	Site example	STRING taxon
Arabidopsis (TAIR10)	`scripts/download_arabidopsis_tair10.sh`	`site_tair10.example.json`	3702
Soybean (Wm82 / v2.1)	`scripts/download_glycine_max_wm82.sh`	`site_glycine_max_wm82.example.json`	3847
Maize (B73 NAM 5.0)	`scripts/download_zea_mays_b73.sh`	`site_zea_mays_b73.example.json`	4577
Wheat (IWGSC)	`scripts/download_triticum_aestivum_iwgsc.sh`	`site_triticum_aestivum_iwgsc.example.json`	4565

scripts/download_arabidopsis_tair10.sh   # or soybean / maize / wheat script
export METHYL_SITE_CONFIG=workflow_engine/domain/profiles/site_tair10.example.json

Wheat toplevel FASTA is large (~4 GB compressed); restrict study chromosomes (e.g. ["1A"]) for research runs if needed.

Instantiate on /work¶

Studies live under /work/projects/<study>/. Scaffold with the CLI, then copy this manifest/overlay in:

methyl-study-init \
  --study-id plant-abiotic-stress \
  --name Control_vs_Drought \
  --analyte plant_tissue \
  --modality methylation \
  --binary \
  --intended-use "Research-use Arabidopsis drought methylation study." \
  --output-root /work/projects
# then fill data/*.csv and assign plant-disjoint validation_partitions.

Run¶

methyl-workflow-run \
  --program workflow_engine/domain/fixtures/plant_stress_study_lifecycle.program.json \
  --context-file docs/examples/samd/plant-abiotic-stress/context_plant_abiotic_stress.json

Enrichment uses the plant-stress-core library preset from the enrichment preset registry (methyl-cfg sync-library-presets), not the oncology cancer-core default. Human-symbol Enrichr libraries do not map to Arabidopsis AGI locus IDs, so for true Arabidopsis/crop term enrichment supply a custom plant GMT via actionConfig.enricher.libraries; plant-stress-core (organism-general GO terms) is the discovery default. Cell deconvolution is not run (blood-only bases); the plant lifecycle program omits that node.

Plant trait prior (not Open Targets). Open Targets / DisGeNET are human-only. The Arabidopsis overlay sets mapper.enrich_source: plant_traits and mapper.plant_traits_path to the demo TSV above (columns: gene, disease_term, optional score / evidence_level). For crops, point plant_traits_path at a SoyBase / MaizeGDB / WheatIS / Gramene extract with the same column names.

Other crops¶

Use the matching project + context pair with the crop site:

export METHYL_SITE_CONFIG=workflow_engine/domain/profiles/site_glycine_max_wm82.example.json
methyl-workflow-run \
  --program workflow_engine/domain/fixtures/plant_stress_study_lifecycle.program.json \
  --context-file docs/examples/samd/plant-abiotic-stress/context_soybean_drought.json

Same program and analyte; only site pins, chromosome list, organism, and STRING taxon change.

Optional: plant cell deconvolution / epi-GBS¶

Atlas: provision a Houseman/HiTIMED JSON, set site actionConfig.cell_deconvolution, run plant_stress_study_lifecycle_with_deconv.program.json (default lifecycle still omits the node). CI fixture: workflow_engine/domain/checks/plant_abiotic_stress/data/plant_houseman_seed_fixture.json.
epi-GBS: libraryProtocol: epi_gbs + sample_prep_epigbs.program.json + actionConfig.docker_align image/argv (Parabricks WGBS SamplePrep stays separate). See ch.23 and plant-deconv-epigbs-seams.