Preprint Seven replicated genomic associations of myalgic encephalomyelitis/chronic fatigue syndrome: a biobank study, 2026, Slaughter, Ponting et al.

[hotblack]

It would be interesting to see if these different genes reflect severity differences, something that we talked about in the Does Very Severe ME/CFS involve an extra process: ME/CFS + X? thread. I think more work would need to be done to verify that, like separating DecodeME’s cohort into severe+v.severe and moderate+mild to see what comes up, or doing a whole other GWAS. I wonder too if there could be a difference in age of onset because we know there was a difference in average age at sample collection, but I don’t know how well the UK Biobank and All of Us recorded that data.

Absolutely agree. If there is any way to do so we should use existing data to look at subgroups within the DecodeME cohort by severity. Sorry to go on about this but it’s come up a lot, we have the data, the barriers seem to be statistical, is we need a better method that works with smaller cohorts. Is targene that? I think it could be.

CSMD1 was a significant gene in two PrecisionLife studies:

Genetic Risk Factors for ME/CFS Identified using Combinatorial Analysis, 2022, Das et al
(In supplementary table 7, disease signature 68.)


Identification of Novel Reproducible Combinatorial Genetic Risk Factors for [ME] in [DecodeME Cohort] and Commonalities with [LC], 2026, Sardell et al
(CSMD1 was one of 259 candidate core genes.)

Yes and in the brain/synapse cluster when I looked at clustering of the PrecisionLife data.

What if a load of different combinations of issues of different brain and immune factors all produce similar results (ME) and the causes are similar but not identical but the output is the same.

That’s enough thinking for one day

We’re all taking different roads to the same destination…

 

[neophyte32]

That’s depressive. Did they give any reason for why? It seems to be a no-brainer to work with the only data points we know point to causal mechanisms.

Does he mention American researchers, particularly researchers at Stanford?

 

[MrMagoo]

Is it difficult to contact the participants again to collect age of onset data?

I would think it’s expensive

 

[Hoopoe]

I would think it’s expensive

It's probably possible to automate the process of sending emails and collecting the responses. It would require some work from someone who knows how to do these things.

 

[MrMagoo]

It's probably possible to automate the process of sending emails and collecting the responses. It would require some work from someone who knows how to do these things.

I know however you would need to pay somebody to do that plus trace the response to the patient’s donation/saliva sample

Otherwise all we would know is of 9,000 samples 30% were under 18 and 20% were 19-30 and so on, but not which ages relate to which data.

So you are looking at funding someone to do that maybe 2 days a week for 6 months at a minimum. That’s probably nearly £10k

 

[Utsikt]

Is it difficult to contact the participants again to collect age of onset data?

You’d not be able to connect the info together if they have anonymised the original data.

Our best bet is probably to push anyone doing any kind of large scale work with ME/CFS to collect age of onset data from now on.

 

[wallfish]

You’d not be able to connect the info together if they have anonymised the original data.

Surely those responsible for the data keep a key somewhere?

 

[Hoopoe]

The data will be stored in a digital database. For each participant, send out an email with a link to a form. The link contains metadata that allows tracing back the response to an individual participant.

The minimum work required is:
Write the script that reads internal participant ID and email and sends an email with a unique link for that participant to the email address.
Set up a form on a website that allows participants to provide the response and store it in a separate database.
Check for duplicate submissions and that everything is okay, then insert the new data into the main database.

Potential problems with this:
I'm pretty sure that DecodeME collected email addresses, but maybe not for all participants.
People might overlook the instruction to not share the link with anyone, sharing it with other patients.

 

[Utsikt]

Surely those responsible for the data keep a key somewhere?

They would need to be able to connect your name to the physical sample so you can request that it’s destroyed, but why would they need to be able to connect the sequencing data to your name?

 

[Andy]

Yes, the DecodeME team kept contact details for all participants, and 95% of all participants gave consent for possible re-contact about future projects.

What I don't believe that we can do is add to the DecodeME dataset in the way that is being proposed here, by recontacting participants to request their age of onset and adding that to their individual dataset. We do not have ethical approval for that, we do not have the staff to do that, and I don't believe that it is even technically possible for us to do it at this stage.

 

[Hoopoe]

@Andy thanks for the clarification. Maybe in a future project a question about age of onset can be added.

 

[V.R.T.]

Yes, the DecodeME team kept contact details for all participants, and 95% of all participants gave consent for possible re-contact about future projects.

What I don't believe that we can do is add to the DecodeME dataset in the way that is being proposed here, by recontacting participants to request their age of onset and adding that to their individual dataset. We do not have ethical approval for that, we do not have the staff to do that, and I don't believe that it is even technically possible for us to do it at this stage.

Could this information be collected about the LC participants of SequenceME &LC?

 

[Andy]

Could this information be collected about the LC participants of SequenceME &LC?

Yes, potentially, as the LC questionnaire is yet to be defined.

 

[V.R.T.]

Yes, potentially, as the LC questionnaire is yet to be defined.

That could be a good idea then imo, because you've got a cohort of people with the same infectious trigger that you can compare data by age of onset in.

 

[wallfish]

They would need to be able to connect your name to the physical sample so you can request that it’s destroyed, but why would they need to be able to connect the sequencing data to your name?

I assumed the data is only anonymized when it is shared. Is the trail from participant to data destroyed/not recorded in studies of this kind?

 

[Utsikt]

I assumed the data is only anonymized when it is shared.

That would be a massive security risk, I would be shocked if they even keep the contact info and genetics in the same storage place.

Is the trail from participant to data destroyed/not recorded in studies of this kind?

I’d imagine you have four types of data:

1. Personal
2. Questionnaires
3. Genetics
4. Biosamples

2 and 3 can be connected for analysis purposes. 1 has to be connectable to 4 to enable you to request that the sample is destroyed.

Connecting across those two groups serve no practical purpose that I can think of for the researchers, so the optimal solution from a risk management perspective is to make it impossible.

It might be that they keep the key to let you remove consent after the sample is processed, but before it has been analysed, but even that isn’t required everywhere. Some don’t even have to destroy samples.

See this as an example (withdrawal from research):

Required Elements of the Consent Form

Explore the basic elements of informed consent that are required by the Common Rule and that are relevant to genomics.

www.genome.gov

 

[Chris Ponting]

Is it difficult to contact the participants again to collect age of onset data?

Yes, drafting and testing a questionnaire, then getting it through ethics and implementing it via Qualtrix (and as a paper based questionnaire) together is a year's work for someone, even before data analysis. Nevertheless, we have not given up hope of delivering a third questionnaire, if we can.

 

[hotblack]

Yes, drafting and testing a questionnaire, then getting it through ethics and implementing it via Qualtrix (and as a paper based questionnaire) together is a year's work for someone, even before data analysis. Nevertheless, we have not given up hope of delivering a third questionnaire, if we can.

Thanks Chris, I’m sure we’d all help spread the news of that if we can!

Are you able to tell us if using the techniques from this paper would enable you to look at subgroups by severity within the DecodeME cohort?

I’m interested in the methodology here. Can we use these approaches and techniques on different datasets or subsets, on DecodeME data?
- We’re comparing different groups with different criteria and different methods at the moment. It is perhaps unsurprising we get different results. And maybe more surprising we got replication with separate cohorts.

- This brings us back to looking at groups by severity imho. This is s something we talked about before on DecodeME, can or should we subgroup by severity? I think the numbers were said to be too small to look at severe only but does this approach change that? Or if not what about excluding severe?

 

[EndME]

"At what age did you first have ME/CFS symptoms?"

My first intuition would be that this might have also been a non-ideal question. Ill people, like everyone else, often try to interpret things retrospectively even if there may have been no connection and as such list a symptom start much earlier than illness onset. To capture both I think one first has to ask something along the lines of "At what age did you develop ME/CFS?" and then if one wants to see if there is something like a prodromal phase something similar to "At what age did your first ME/CFS symptoms become apparent?"

 

[V.R.T.]

My first intuition would be that this might have also been a non-ideal question. Ill people, like everyone else, often try to interpret things retrospectively even if there may have been no connection and as such list a symptom start much earlier than illness onset. To capture both I think one first has to ask something along the lines of "At what age did you develop ME/CFS?" and then if one wants to see if there is something like a prodromal phase something similar to "At what age did your first ME/CFS symptoms become apparent?"

Yes, I find a simple age of onset question tricky to answer because I had a clear infectious 'definite' onset at age 26 but something like prodromal symtoms from around age 19. If I'm asked a simple at what age question I say 26, but that answer would miss the fact that I stopped being 'healthy' and developed a sleep disorder and other common symtoms of MECFS close to the first age peak, even if I didn't unquestionably develop ME/CFS with clear PEM for years afterwards.

So I agree that the best version of asking about age of onset allows for people to give an answer that allows for a prodromal phase to be noted.