Preprint Initial findings from the DecodeME genome-wide association study of myalgic encephalomyelitis/chronic fatigue syndrome, 2025, DecodeMe Collaboration

I have a question about the gene sequencing reference used in the study. I thought the GRCh37 (hg19) reference was used to interpret the GWAS gene array data in the original UK Biobank work. The DecodeME study uses the same technology so that comparison to the original data for controls is valid.

However the gene reference to the variants listed in the paper seems to be using the GRCh38 (hg38) reference. That means if we want to compare a variant location using the UK Biobank online tool we have to map the coordinates. For example, OLFM4 variant 13-53194927-GT-G is a GRCh38 reference that maps to 13-53769062-GT-G in GRCh37. That seems to map to rs35306732.

When I search for that variant in the CFS phenotype in the UK Biobank it doesn't show any results for that variant. Am I going wrong somewhere? Surely it should be in the original database?

Search | GeneATLAS

geneatlas.roslin.ed.ac.uk geneatlas.roslin.ed.ac.uk
 
wigglethemouse said:
Has there been any comment of the previous finding of SLC25A15 in women in the UK Biobank ME/CFS dataset highlighted in the Joshua Dibble thesis, a member of Chris Pontings team. I couldn't find a reference in the new paper. Would be interesting to know the p-value in this study.
Click to expand...

I checked the summary stats for the variant from the thesis for SLC25A15, rs7337312, which has the ID 13:40779161:G:A. The p-value for this in DecodeME is 0.713705.

Dibble found another novel association in males with the gene PDE10A, with the SNP rs76346913, which has the ID 6:165736174:C:T, but that variant doesn't seem to be one of the variants they tested. I think you would then do something like look up which SNPs are in LD with this SNP to see if any of them are in DecodeME to see if they are significant, but I don't know enough to know how to do that right.

Edit: The thesis does provide a SNP in strong LD with the PDE10A SNP, rs78375762:
However, to support this as a true association, it is encouraging that there are two other neighbouring SNPs on chromosome 6 with p-values near the suggestive threshold for this phenotype in males: rs1013902 and rs78375762. rs1013902 has the more notable p-value (p = 2.20 × 10−6 ), but is not in LD with rs76346913 (D0 = 0.0203). However the second SNP (rs78375762, p = 6.70 × 10−6 ) is in strong linkage disequilibrium with rs76346913 (D0 = 0.912).
Click to expand...
The p-value for this second SNP in DecodeME is 0.98 in males, 0.83 in the full cohort.

Edit: And last one I checked, Dibble also found a near significant variant associated with the gene LPGAT1: rs116109664. The p-value for this variant in DecodeME is 0.28.
 
Last edited:
forestglip said:
Dibble found another novel association in males with the gene PDE10A, with the SNP rs76346913, which has the ID 6:165736174:C:T, but that variant doesn't seem to be one of the variants they tested.
Click to expand...
It's puzzling why we can't find matches between the variants in the two sets of data. i wouldn't have expected to find multiple missing data in only a couple of variants we have spot checked for comparison..
 
wigglethemouse said:
It's puzzling why we can't find matches between the variants in the two sets of data. i wouldn't have expected to find multiple missing data in only a couple of variants we have spot checked for comparison..
Click to expand...
I'm not sure about the one in your previous post. I would expect it to be in the BioBank. But maybe that website GeneAtlas doesn't show every variant they tested for whatever reason.

For the one that Dibble tested in the BioBank that wasn't tested in DecodeME, maybe that's just a variant that didn't get a good read in the DecodeME sequencing.

Edit: For the variant that isn't in GeneAtlas, I see it on this other website that just shows the allele frequency in the BioBank: https://afb.ukbiobank.ac.uk/variant/chr13-53194927-GT-G Maybe it just wasn't compared to CFS in the BioBank for some reason. But the data is still there for a few hundred thousand people.
 
Last edited:
voner said:
was anyone else besides me surprised that 86.1% of the cohort reported "muscle pain"? many years of following ME/CFS forums, webinars, conferences led me to believe the number would be much lower. I’m not saying this number is good or bad or inaccurate, just that it surprised me.
Click to expand...
I was surprised at the significance of pain too.
I do get a lot of pain but linked I thought mostly to musculoskeletal pain. Also pain from lipomas. Also spasmodic pain around shoulder blades. I have h(EDS). For the last 18 years I have been taking prescribed pain relief which might hide other types of pain but I was still surprised at this result.
 
Regarding long covid, there are a few mentions of long covid and COVID-19 itself in the "Candidate Gene" document Chris Ponting shared in post #15 upthread. I've copied the info on each candidate gene and bolded mentions of long covid and COVID itself.

SERPINC1 (Tier 1)
• Protein: Antithrombin-III. UniProt. GeneCards. The allele that increases the risk of ME/CFS is associated with decreasing SERPINC1 gene expression.
• Molecular function: Serine protease inhibitor in plasma that regulates the blood coagulation cascade: AT-III inhibits thrombin and matriptase-3/TMPRSS7, as well as factors IXa, Xa and XIa.
• Cellular function: Extracellular function.
• Link to disease: Mutations in SERPINC1 cause antithrombin-III deficiency, which is a strong risk factor for thrombosis. A reduction in the serum level of this protein is associated with severe cases of COVID-19. Mutated in cerebral venous sinus thrombosis (CVST).
• Potential relevance to ME/CFS: Previously proposed role of ischaemia–reperfusion injury in chronic, relapsing diseases such as rheumatoid arthritis, Long Covid, and ME/CFS (23). Reduction in SERPINC1 levels expected to increase risk of blood clots. However, the eQTL is only found in brain tissue, where it is only expressed at a low level.
Click to expand...

BTN2A2 (Tier 1)
• Protein: Butyrophilin subfamily 2 member A2. UniProt. GeneCards. The allele that increases the risk of ME/CFS is associated with decreasing BTN2A2 gene expression.
• Molecular function: Mouse Btn2a2 inhibits T-cell activation and promotes the induction of regulatory T cells (Tregs) (31). It also promotes central tolerance to prevent autoimmune disease (32).
• Cellular function: Innate immunity, microbe-sensing.
• Link to disease: Butyrophilins confer immunity against diverse pathogenic bacteria and parasites.
• Potential relevance to ME/CFS: Stoichiometric imbalance of BTN2A and BTN3A levels could alter the ability of Vγ9Vδ2 T cells to discriminate microbial- from self-antigens, and to kill infected cells. Vγ9Vδ2 T cells can be activated by TLR3 and TLR4 ligands. TLR4 variants have been associated with Long Covid in a combinatorial genetics analysis (33).
Click to expand...

BTN3A3 (Tier 1)
• Protein: Butyrophilin subfamily 3 member A3. UniProt. GeneCards. The allele that increases the risk of ME/CFS is associated with increased BTN3A3 expression in skin not exposed to the sun.
• Molecular function: Co-expression of BTN3A3 synergises with BTN3A1 to greatly boost the efficiency of pAg sensing (42). Confers a chaperone-like function in boosting cell surface expression of BTN3A1 (43).
• Cellular function: Innate immunity, microbe-sensing.
• Link to disease: Butyrophilins confer immunity against diverse pathogenic bacteria and parasites.
• Potential relevance to ME/CFS: Stoichiometric imbalance of BTN2A and BTN3A levels could alter the ability of Vγ9Vδ2 T cells to discriminate microbial- from self-antigens, and to kill infected cells. Vγ9Vδ2 T cells can be activated by TLR3 and TLR4 ligands. TLR4 variants have been associated with Long Covid in a combinatorial genetics analysis (33).
Click to expand...

This one is very long so I have not pasted everything. See the candidate gene document for more details:
ZNFX1 (Tier 1)
• Protein: Zinc finger NFX1-type containing 1. UniProt. GeneCards. The allele that increases the risk of ME/CFS is associated with increased ZNFX1 expression in blood and brain tissues.
• Molecular function: An interferon-stimulated dsRNA sensor that specifically restricts the replication of RNA viruses early after infection (82). Studies are unclear whether ZNFX1 is localised to the mitochondrion.
...
Among 3,044 COVID-19 cases, ZNFX1 expression level shows a strong negative correlation with SARS-CoV-2 viral load (85).
Click to expand...
 
Re: chronic pain, how is this defined? What does it mean in the context of genetics results? I see there have been multiple GWAS papers on chronic pain, but it's surprising to me that people with pain of various kinds might have a common signal genetically? There are so many conditions and situations that can cause chronic pain. Sorry if this is too basic a question.

I do have pain personally but I don't know if I would describe it as in my "muscles". Worryingly I can't remember what I answered in the questionnaire. I suppose I would answer yes, but most of the pain feels like it floats around independently of my muscles.
 
Ariel said:
Re: chronic pain, how is this defined? What does it mean in the context of genetics results? I see there have been multiple GWAS papers on chronic pain, but it's surprising to me that people with pain of various kinds might have a common signal genetically? There are so many conditions and situations that can cause chronic pain. Sorry if this is too basic a question.

I do have pain personally but I don't know if I would describe it as in my "muscles". Worryingly I can't remember what I answered in the questionnaire. I suppose I would answer yes, but most of the pain feels like it floats around independently of my muscles.
Click to expand...
Previously I wondered if the “Chronic pain gene“ (CA10 I think?), might actually be associated with ME not chronic pain. But since ME is commonly undiagnosed and has chronic pain for many, a lot of pwME may have ended up in a chronic pain GWAS and with large sample sizes it might show an effect.
 
Yann04 said:
Previously I wondered if the “Chronic pain gene“ (CA10 I think?), might actually be associated with ME not chronic pain. But since ME is commonly undiagnosed and has chronic pain for many, a lot of pwME may have ended up in a chronic pain GWAS and with large sample sizes it might show an effect.
Click to expand...
Yes, I was thinking the same thing. I was quite taken aback at all the genetic research for "chronic pain" compared to what we have been getting research-wise. I imagine this research looked at quite different cohorts but I am not well enough to read about it right now. Maybe someone who understands better how the analysis was done re: the chronic pain comparison will clarify the situation.
 
Yann04 said:
Previously I wondered if the “Chronic pain gene“ (CA10 I think?), might actually be associated with ME not chronic pain. But since ME is commonly undiagnosed and has chronic pain for many, a lot of pwME may have ended up in a chronic pain GWAS and with large sample sizes it might show an effect.
Click to expand...
That's an interesting thought.

Conversely, there seems to be a lot of overlap between ME and Fibro. Maybe the people who have this chronic pain gene
/have large amounts of pain with their ME symptoms - have fibro+ME.
 
Yann04 said:
Previously I wondered if the “Chronic pain gene“ (CA10 I think?), might actually be associated with ME not chronic pain. But since ME is commonly undiagnosed and has chronic pain for many, a lot of pwME may have ended up in a chronic pain GWAS and with large sample sizes it might show an effect.
Click to expand...
How does that make sense? If these people with "the same ME/CFS as those in DecodeME" end up in chronic pain cohort then shouldn't the chronic pain study find also the other genes from DecodeME?
 
V.R.T. said:
That's an interesting thought.

Conversely, there seems to be a lot of overlap between ME and Fibro. Maybe the people who have this chronic pain gene
/have large amounts of pain with their ME symptoms - have fibro+ME.
Click to expand...
Anecdotally I know of more than one case where a person with ME/CFS partially "recovered" from that and their illness turned into fibromyalgia with a pain presentation basically. Interestingly the one case I am thinking of in particular involves a person where pain medication eg opiates, whatever they give you at the dentist (local anaesthetic), does not work on them - apparently this is very rare and for genetic reasons? I thought that was interesting that they have so much pain and can't really get pain relief that way. (They don't work at all seemingly for them). A fairly cruel coincidence?
 
Ariel said:
Thanks; I would be curious about this because I am a participant and one of my parents did immigrate but one did not. Presumably this means I would be included but I cannot remember what questions I answered. :( I don't remember the questions being asked or what I wrote.

I am not sure in general though - how long does one's family have to be here to have not "immigrated"? I don't understand this criterion. I get it about comparing similar genes but it would be good to know if I was actually included in the GWAS portion of the study and what this means for others.
Click to expand...
Yes I was wondering this, because I have inter generational mixed-race, it’s recent enough for me to have had a sickle cell test but I don’t recall what I was asked about on the DecodeME.
 
EndME said:
How does that make sense? If these people with "the same ME/CFS as those in DecodeME" end up in chronic pain cohort then shouldn't the chronic pain study find also the other genes from DecodeME?
Click to expand...
Not necessarily. As we saw in decode the effects are quite marginal. Perhaps as @VRT suggests CA10 is associated with more Fibro adjacent presentations of ME, ie. more likely to end up in chronic pain cohorts.

I think we shouldn’t assume ME as “one illness”, yet.
 
I remember struggling a little with that question as on pain meds and I couldn't remember properly what drove me onto them in the first place. But I ticked yes in the end as I do get what probably are a muscular type aches in my legs, not exactly pain but that make me "squirm" and I have to contract the muscles to try and relieve the sensation. It's quite widespread so I hesitated at first to define it as muscular. I definitely don't have Fibromyalgia. I only get joint pain in my elbows in extreme pem, when not on meds worsened in winter.
 
Yann04 said:
Not necessarily. As we saw in decode the effects are quite marginal. Perhaps as @VRT suggests CA10 is associated with more Fibro adjacent presentations of ME, ie. more likely to end up in chronic pain cohorts.

I think we shouldn’t assume ME as “one illness”, yet.
Click to expand...
But isn't this then just the argument that is already presented? That genes related to pain can be part of different illnesses associated to pain in various ways?
 
One thing as a non technical lay person I still don't understand as how the 8 signals work in an individual.

If you have ME do you only need one of them, so there are 8 phenotypes?

Or did you need two or three? Or all 8?

Thanks if someone who can explain how that bit works, I can't read the technical paper.
 
You must log in or register to reply here.
Back
Top Bottom