DecodeME Genome-Wide Association Study

1 DecodeME Consortium 2025 — The Largest ME/CFS GWAS

Full Citation:: DecodeME Consortium, Ponting CP, et al. Initial findings from the DecodeME genome-wide association study of myalgic encephalomyelitis/chronic fatigue syndrome. medRxiv. 2025. Preprint v2, August 2025. DOI:: 10.1101/2025.08.06.25333109v1 Study Design:: Genome-wide association study (GWAS); case-control Sample Size:: >15,000 ME/CFS patients; population controls from UK Biobank and other cohorts Key Findings::

- 8 million+ SNPs tested; 8 genome-wide significant loci identified ($p < 5 \times 10^{-8}$)
- Strongest signal: chr20 ($-log_10 p = 11.02$, OR = 1.095); candidate genes *ARFGEF2*, *CSE1L*, *STAU1*
- Other loci implicate *CA10* (chr17), *UNC13C* (chr15), *OLFM4* (chr13)
- All 8 loci are common variants (MAF 13--63%) with small effect sizes (OR 0.93--1.10)
- SNP heritability: $h^2_\text{SNP} = 0.095$ (9.5%) via LD Score Regression
- MAGMA tissue expression analysis: all significantly enriched tissues are brain regions
- Genetic correlations: IBS ($r_g = 0.75$), depression ($r_g = 0.60$), fatigue/sleep traits ($r_g = 0.57$--$ 0.66$)
- No significant genetic correlation with MS, RA, Crohn's disease, T1D, or T2D
- Genomic inflation factor: 1.066 (minimal population stratification)

Relevance:: Watershed study establishing the polygenic, brain-focused genetic architecture of ME/CFS. The convergence on neuronal genes challenges purely peripheral/immune models. The absence of autoimmune genetic correlation constrains the autoimmune hypothesis. The autophagy gene signals (FBXL4, CCPG1) provide a novel genetic link to mitochondrial dysfunction. Effect sizes are small individually but the loci point to major biological pathways amenable to therapeutic targeting. Certainty Assessment::

- *Quality:* High (largest ME/CFS genetic study ever; rigorous methodology; British ancestry with PCA correction)
- *Replication:* Not yet replicated (by design---discovery cohort; no comparable replication cohort exists)
- *Certainty:* 0.70 (robust methodology and convergence with independent rare-variant study; awaits independent replication and peer review)

2 ME/CFS Science 2025 — Critical Analysis of DecodeME (Two-Part Series)

Full Citation:: ME/CFS Science. DecodeME: the biggest ME/CFS study ever. mecfsscience.org. October 4, 2025. Part II: Genes pointing to the brain: DecodeME part II. October 13, 2025. URLs:: Part I | Part II Source Type:: Critical analysis blog by two long-term ME/CFS patients; focus on statistical methodology and evidence evaluation Key Contributions::

- Part I: Contextualises effect sizes (1--2 percentage point prevalence differences); compares SNP heritability across conditions; presents full genetic correlation table against 3,167 UK Biobank traits; explains why small GWAS effects can point to major therapeutic targets (HMGCR/statin example)
- Part II: Fine-mapping analysis identifying specific genes at each locus via GTEx, proximity, and GWAS Catalog cross-referencing; three biological themes (neuronal, immune-ambiguous, autophagy); convergence with Stanford rare-variant study (*NLGN2*, *SYNGAP1*); confounding analysis (symptom overlap, intelligence/education); GWAS Catalog prior associations for each gene

Relevance:: Provides the most accessible and rigorous public analysis of DecodeME findings. The gene-by-gene characterisation and cross-referencing with prior GWAS is not available in the preprint itself. Critical for understanding what DecodeME does and does not tell us about ME/CFS biology. Certainty Assessment::

- *Quality:* High (thorough statistical reasoning; transparent about limitations)
- *Certainty:* N/A (secondary analysis; certainty pertains to the primary data)