1) This new study looked at the association between genetics and metabolism in both ME/CFS and controls using the UK Biobank. Several genes (LINC01701, HSD11B1, SCGN, and NLRC5) were significantly associated with metabolic markers in ME/CFS only.
2) This comes from the research group of metabolomics expert Christopher Armstrong. They used data on 135 NMR metabolic biomarkers, such as lipids and fatty acid ratios, and looked at which genes are associated with these metabolites in the ME/CFS and control groups separately. 3) They found 27 significant SNPs that mapped to 15 genes in the ME/CFS group and many more in the control group, which was much bigger (875 ME/CFS patients compared to 36,033 controls). 4) The most interesting part is where they compared the gene associations in both groups. The graph below shows the effect sizes for in both the control and ME/CFS group, and the genes were there was a clear difference.
5) Four genes were only significantly associated with metabolic markers in the ME/CFS group. These include HSD11B1 and SCGN, which have roles along the HPA-axis (which came up in a recent Dutch autopsy report on 7 ME/CFS cases). 6) Another notable gene was NR1H3, which encodes liver X receptor alpha (LXRalpha). The study suggested that these metabolic pathways may play a different role in ME/CFS than in controls. 7) There are some big caveats though. These four genes were not associated with ME/CFS in DecodeME and other big genetic databases (see Table 2 below). The link between a DNA letter and a gene is also often quite uncertain (as analyses of DecodeME showed).
8) Another issue is that the comparison of effect sizes was complicated, as the sample size in ME/CFS was much smaller. This creates more spurious big effect sizes than in larger sample sizes, known as 'winner's curse' (the authors tried to correct for this bias, though). 9) The link to the paper: Huang et al. 2025. Exploring a genetic basis for the metabolic perturbations in ME/CFS using UK Biobank. sciencedirect.com