{
  "schema_version": "2.0.0",
  "harmonisation_ruleset": "pais-harmony-v1",
  "id": "decodeme-mecfs-gwas",
  "name": "DecodeME ME/CFS GWAS",
  "aliases": ["DecodeME genome-wide association study"],
  "outbreak_event": null,
  "pathogen_id": "unknown-trigger",
  "pathogen_class": "unknown",
  "exposure_ascertainment": "not_reported",
  "design": "registry_linkage",
  "denominator_defined": "yes",
  "denominator_method": "Genotyped case-control sample: 15,579 people with ME/CFS vs 259,909 population controls of European descent, recruited via a dedicated UK study plus biobank linkage (UK Biobank, Lifelines).",
  "recruitment_source": "population_registry",
  "time_zero_definition": "Germline genotype (fixed at conception); ME/CFS caseness ascertained cross-sectionally",
  "time_zero_precision": "undefined",
  "control_group": "unexposed_unmatched",
  "control_matching_variables": ["genetic ancestry"],
  "n_source_population": null,
  "n_enrolled": 15579,
  "n_analysed": 15579,
  "followups": [],
  "max_followup_months": null,
  "instruments": ["dsq"],
  "case_definition": "author_defined",
  "pem_assessed": "yes_single_item",
  "objective_measures": ["none"],
  "specimens_collected": true,
  "specimen_types": ["saliva"],
  "acute_phase_specimens": "no",
  "storage": null,
  "biobank_status": "active",
  "consent_future_use": "broad",
  "external_access": "collaboration_only",
  "contact": null,
  "registration_id": null,
  "funders": ["Medical Research Council (UK)", "National Institute for Health and Care Research (NIHR)"],
  "conflicts": null,
  "data_availability": "Summary statistics released by the DecodeME collaboration.",
  "notes": "The largest ME/CFS genome-wide association study (15,579 cases). Included as the anchor host-genetic susceptibility cohort for an idiopathic syndrome: eight genome-wide signals, of which four (RABGAP1L, FBXL4, OLFM4, CA10) replicated in independent post-exertional-malaise/fatigue samples. Genetic factors are temporality-valid by construction (fixed at conception). Initial findings are a 2025 preprint — flagged accordingly.",
  "symptom_inventory_scope": "none",
  "n_symptoms_queried": null,
  "symptom_instrument_note": null,
  "related_cohorts": [
    { "id": "uk-me-cfs-biobank", "relation": "sibling_analysis" },
    { "id": "nih-intramural-me-cfs-2024", "relation": "sibling_analysis" }
  ],
  "flags": ["preprint", "not_peer_reviewed"],
  "last_verified": "2026-07-20",
  "verified_by": "OSMF research-tracker (web-verified against source)",
  "publications": [
    {
      "id": "decodeme-2025",
      "title": "Initial findings from the DecodeME genome-wide association study of myalgic encephalomyelitis/chronic fatigue syndrome",
      "authors": "DecodeME Collaboration.",
      "year": 2025,
      "journal": "medRxiv (preprint)",
      "type": "preprint",
      "doi": "10.1101/2025.08.06.25333109",
      "url": "https://www.medrxiv.org/content/10.1101/2025.08.06.25333109v1",
      "open_access": true,
      "is_primary_cohort_paper": true
    }
  ],
  "observations": [],
  "predictors": [
    {
      "id": "decodeme-pred-rabgap1l",
      "cohort_id": "decodeme-mecfs-gwas",
      "publication_id": "decodeme-2025",
      "question_type": "conversion",
      "factor_id": "gene:rabgap1l",
      "factor_verbatim": "RABGAP1L locus (genome-wide significant signal)",
      "measurement_window": "genetic_fixed",
      "factor_value_type": "binary",
      "contrast": "risk allele vs reference at the RABGAP1L signal",
      "reference_level": "reference allele",
      "outcome_measure_id": "case-def:cfs",
      "outcome_timepoint_months": null,
      "estimate": { "type": "none", "value": null, "ci_low": null, "ci_high": null, "p_value": null, "direction": "increases_risk", "significant": true },
      "model": { "type": "mixed_effects", "adjusted_for": ["genetic principal components"], "prespecified": "yes", "n_predictors_tested": null, "multiplicity_correction": "other" },
      "n_analysed": 15579,
      "analysis_status": "analysed",
      "harmonised": null,
      "provenance": { "source_locator": "Results (one of eight genome-wide signals; replicated in PEM/fatigue samples)", "extracted_by": "OSMF", "extracted_on": "2026-07-20", "verified_by": "OSMF research-tracker", "verified_on": "2026-07-20", "extraction_method": "manual" }
    },
    {
      "id": "decodeme-pred-fbxl4",
      "cohort_id": "decodeme-mecfs-gwas",
      "publication_id": "decodeme-2025",
      "question_type": "conversion",
      "factor_id": "gene:fbxl4",
      "factor_verbatim": "FBXL4 locus (genome-wide significant signal)",
      "measurement_window": "genetic_fixed",
      "factor_value_type": "binary",
      "contrast": "risk allele vs reference at the FBXL4 signal",
      "reference_level": "reference allele",
      "outcome_measure_id": "case-def:cfs",
      "outcome_timepoint_months": null,
      "estimate": { "type": "none", "value": null, "ci_low": null, "ci_high": null, "p_value": null, "direction": "increases_risk", "significant": true },
      "model": { "type": "mixed_effects", "adjusted_for": ["genetic principal components"], "prespecified": "yes", "n_predictors_tested": null, "multiplicity_correction": "other" },
      "n_analysed": 15579,
      "analysis_status": "analysed",
      "harmonised": null,
      "provenance": { "source_locator": "Results (one of eight genome-wide signals; replicated)", "extracted_by": "OSMF", "extracted_on": "2026-07-20", "verified_by": "OSMF research-tracker", "verified_on": "2026-07-20", "extraction_method": "manual" }
    },
    {
      "id": "decodeme-pred-olfm4",
      "cohort_id": "decodeme-mecfs-gwas",
      "publication_id": "decodeme-2025",
      "question_type": "conversion",
      "factor_id": "gene:olfm4",
      "factor_verbatim": "OLFM4 locus (genome-wide significant signal)",
      "measurement_window": "genetic_fixed",
      "factor_value_type": "binary",
      "contrast": "risk allele vs reference at the OLFM4 signal",
      "reference_level": "reference allele",
      "outcome_measure_id": "case-def:cfs",
      "outcome_timepoint_months": null,
      "estimate": { "type": "none", "value": null, "ci_low": null, "ci_high": null, "p_value": null, "direction": "increases_risk", "significant": true },
      "model": { "type": "mixed_effects", "adjusted_for": ["genetic principal components"], "prespecified": "yes", "n_predictors_tested": null, "multiplicity_correction": "other" },
      "n_analysed": 15579,
      "analysis_status": "analysed",
      "harmonised": null,
      "provenance": { "source_locator": "Results (one of eight genome-wide signals; replicated)", "extracted_by": "OSMF", "extracted_on": "2026-07-20", "verified_by": "OSMF research-tracker", "verified_on": "2026-07-20", "extraction_method": "manual" }
    },
    {
      "id": "decodeme-pred-ca10",
      "cohort_id": "decodeme-mecfs-gwas",
      "publication_id": "decodeme-2025",
      "question_type": "conversion",
      "factor_id": "gene:ca10",
      "factor_verbatim": "CA10 locus (genome-wide significant signal; colocalises with multisite chronic pain)",
      "measurement_window": "genetic_fixed",
      "factor_value_type": "binary",
      "contrast": "risk allele vs reference at the CA10 signal",
      "reference_level": "reference allele",
      "outcome_measure_id": "case-def:cfs",
      "outcome_timepoint_months": null,
      "estimate": { "type": "none", "value": null, "ci_low": null, "ci_high": null, "p_value": null, "direction": "increases_risk", "significant": true },
      "model": { "type": "mixed_effects", "adjusted_for": ["genetic principal components"], "prespecified": "yes", "n_predictors_tested": null, "multiplicity_correction": "other" },
      "n_analysed": 15579,
      "analysis_status": "analysed",
      "harmonised": null,
      "provenance": { "source_locator": "Results (genome-wide signal near CA10; colocalises with chronic pain)", "extracted_by": "OSMF", "extracted_on": "2026-07-20", "verified_by": "OSMF research-tracker", "verified_on": "2026-07-20", "extraction_method": "manual" }
    }
  ],
  "estimates_status": "none"
}