SugarSquared
Senior Member (Voting Rights)
I registered for this a couple weeks ago! I’m very excited!
SequenceME genetic study - from Oxford Nanopore Technologies, the University of Edinburgh and Action for ME
Jonathan Edwards
Senior Member (Voting Rights)
Very good webinar. Things are moving and having got this far I see no way that they will not get the resources to finish.
It is really remarkable to see a community based research programme like this reporting in public. It is quite detailed stuff, but all the time the focus is on helping people to understand why this is such a good project and how important the patient community has been.
Chris mentioned how impressed he was by citizen scientist contributions to analysing the DecodeME data. The whole community has something to offer here. And the opportunities are going to open up more with time.
It is really remarkable to see a community based research programme like this reporting in public. It is quite detailed stuff, but all the time the focus is on helping people to understand why this is such a good project and how important the patient community has been.
Chris mentioned how impressed he was by citizen scientist contributions to analysing the DecodeME data. The whole community has something to offer here. And the opportunities are going to open up more with time.
SugarSquared
Senior Member (Voting Rights)
I agree, what an awesome webinar! The explanations were accessible to non-scientists like me, but they also felt complete. I thought I knew everything about SequenceME, but it turns out I didn't! I had never heard of structural changes in DNA. Very exciting stuff.
It was nice to see and hear you Andy, as well as Chris and Sonya.
For all who missed it, SolveME said they will post this webinar in 48 hours.
It was nice to see and hear you Andy, as well as Chris and Sonya.
For all who missed it, SolveME said they will post this webinar in 48 hours.
Chris was saying something like realistically, he doesn't expect more than around a third of patients to be helped by the initial WGS research. I don't remember the exact wording.
I don't see how we can estimate the proportion helped before we have any idea of specific mechanisms. It's always possible that PEM has a shared core mechanism in everyone, and I would think that the variants could potentially point to treatments that help everyone.
I don't see how we can estimate the proportion helped before we have any idea of specific mechanisms. It's always possible that PEM has a shared core mechanism in everyone, and I would think that the variants could potentially point to treatments that help everyone.
Some posts have been moved to: Spironolactone for ME/CFS in a Patient Homozygous for rs5522 (I180V): A Case Report, 2026, Donnellan et al
Last edited:
mariovitali
Senior Member (Voting Rights)
Unfortunately I couldn't post the following question, it would be great if someone could confirm the following :
DecodeME has already shown that ME/CFS has a common-variant polygenic component. SequenceME is less likely to outperform DecodeME for discovering many tiny common-variant effects, but it could still be very valuable for resolving causal genes, rare/structural variation, biological pathways, and disease subtypes.
Is the above correct?
DecodeME has already shown that ME/CFS has a common-variant polygenic component. SequenceME is less likely to outperform DecodeME for discovering many tiny common-variant effects, but it could still be very valuable for resolving causal genes, rare/structural variation, biological pathways, and disease subtypes.
Is the above correct?
Utsikt
Senior Member (Voting Rights)
I think SequenceME will look at all 3 billion genes, not just the one million where changes are the most common.
My impression was that this would provide more accuracy, not less. The experts might tell me I’m wrong.
Sequence ME & Long Covid - ME Genetics Centre of Excellence
Sequence ME & Long Covid A groundbreaking partnership is working together to secure funding for a study to analyse the […]
megenetics.org.uk
Andy
Senior Member (Voting rights)
From the FAQs, available here
"How is this study different to DecodeME?
DecodeME used a genome-wide association study (GWAS) to identify common genetic variants associated with ME/CFS susceptibility. It found 8 genomic regions linked to the condition.
Sequence ME & Long Covid builds on this work by using long-read whole genome sequencing. This approach is a thousand-times more detailed than GWAS. It identifies: rare genetic variants that are not usually captured by GWAS, structural variation in DNA, such as insertions, deletions and repeat expansions and other complex regions of the genome that are difficult to resolve with standard GWAS methods. The Oxford Nanopore technology also indicates chemical modifications of DNA."
Jonathan Edwards
Senior Member (Voting Rights)
I am not sure that DecodeME points to a polygenic interaction in causation. Each of these risk genes may tip individuals towards getting ME/CFS independently. With a total genetic component of around 10-20% it may be that for any one person only one of these eight genes, or indeed none, was relevant to their developing disease.
I don't think finding more risk genes is in itself very interesting in terms of filling up an overall causal picture. Where I think it may be extremely valuable is in finding specific gene variants that confer very major risk because they are likely to bring in to much sharper focus what the critical pathways are.
In lupus I suspect that most of the genetic component is HLA based. The link to HLA gives us a confirmation of the broad idea that it involves an adaptive immune response. However, the near 100% risk conferred by very rare homozygous C1q deficiency indicates a crucial role for complement malfunction, which leads us to a much more detailed understanding of the disease. The same applies to the link to rare TLR-7 defects. The problem is all about B cells being confused by the balance of co-factor signals mediated by complement and TLR-7.
The Zhang paper has already pointed to some rare genes but at present I don't think we have the confidence in those data that we can have in DecodeME because the methodology is more opaque.
In a way all we need is one rare gene variant that points to a gene we haven't even thought of being relevant but which, if it has to be included, which the data might indicate, completely re-focuses the pathway analysis. Ideally that would be a re-focus that allows everything else so far to slot into place, if a slightly different place from what we have considered.
SugarSquared
Senior Member (Voting Rights)
Posting this here from another thread about Action for ME’s webinar. Slides are likely the same, so if people missed the presentation today from SolveME, you can watch this video while waiting for SolveME’s to be posted.
I really appreciate all of the science communication from the Decode and Sequence teams.
hotblack
Senior Member (Voting Rights)
For people interested it’s worth looking at some of Oxford Nanopore’s blurb, one of the things I’m interested in beyond the core sequencing is methylation.
There’s a lot of info linked from here
nanoporetech.com
Including an overview of the technology
nanoporetech.com
And their documentation page gies into more detail
Particularly this read on human genomic variant and methylation analysis with the platform
nanoporetech.com
My understanding is that this will be able to tell us not just about genetic differences/variants with a huge level of detail (better than could be done in the past) but also give some insight into epigenetic changes (although this may be cell dependant). More info here
nanoporetech.com
I don’t understand this all but it looks like a huge amount of information will be available to analyse. I seem to remember petabytes of data being talked about.
There’s also various short snazzy videos on their YouTube page.
There’s a lot of info linked from here
Oxford Nanopore Technologies Platform Solution
Discover the benefits of nanopore sequencing and find out which sequencing device is right for your output requirements.
nanoporetech.com
How nanopore sequencing works
Oxford Nanopore has developed a new generation of DNA/RNA sequencing technology. It is the only sequencing technology that offers real-time analysis (for rapid insights), in fully scalable formats from pocket to population scale, that can analyse native DNA or RNA and sequence any length of fragment
nanoporetech.com
And their documentation page gies into more detail
Particularly this read on human genomic variant and methylation analysis with the platform
Workflow overview: human variant calling
This end-to-end workflow provides a scalable method to identify previously hidden and potentially pathogenic variants.
nanoporetech.com
My understanding is that this will be able to tell us not just about genetic differences/variants with a huge level of detail (better than could be done in the past) but also give some insight into epigenetic changes (although this may be cell dependant). More info here
Epigenetics and methylation analysis
Directly sequence DNA and RNA modifications with PCR-free nanopore sequencing – no additional library preparation steps needed. Characterise and phase long-range epigenetic changes, structural variants, and single nucleotide polymorphisms in a single dataset with any-length nanopore reads.
nanoporetech.com
There’s also various short snazzy videos on their YouTube page.
Last edited:
hotblack
Senior Member (Voting Rights)
The webinar looks to have been posted
Jonathan Edwards
Senior Member (Voting Rights)
I may have misunderstood but I think Chris said that there was a possibility that gene variants picked up on DecodeME were contributing to other, co-morbid, conditions rather than ME/CFS. I think he mentioned anaemia and depression.
I am a bit puzzled by this. My understanding is that any gene variant picked up on DecodeME as correlated with ME/CFS must be causal for ME/CFS. It may be involved in a pathway that also leads via various branches to a co-morbidity but that does not detract from the causal relation to ME/CFS.
That is, assuming that there really are 'co-morbidities' for ME/CFS. That term is often linked to EDS and MCAS, which probably are not com-morbidities. I am not sure we have good evidence for any co-morbidities.
Reactive depression occurs in people with ME/CFS but it does with most chronic illness and probably has nothing to do with spontaneous depressive illness or what is often called Major Depressive Disorder, which is what genetic studies tend to look at. I would not call reactive depression a co-morbidity.
I was unaware of any link to anaemia and I doubt there is one. There might be dietary deficiencies due to food intolerance in ME/CFS but if they led to anaemia that would not be an independent 'co-morbidity' with a separate genetic risk base.
What I see as much more likely is that some common clinical problems like anaemia, insulin resistance or hyperlipidemia have turned up as increased in ME/CFS cohorts for reasons of bias in cohort selection (Hutan has discussed this in detail). They are co-morbidities with ME/CFS as defined by those cohorts that tend to get recruited for ME/CFS studies but that is not what we want ME/CFS to be defined as. We want to define it as a putative common pathogenic pathway and clinical presentation that will not necessarily be ideally represented by cohorts that get recruited.
I may have got this wrong but I think it would be worth trying to get it clear in our minds. I don't see variants from DecodeME as being separable off as 'causing co-morbidities' rather than ME/CFS.
I am a bit puzzled by this. My understanding is that any gene variant picked up on DecodeME as correlated with ME/CFS must be causal for ME/CFS. It may be involved in a pathway that also leads via various branches to a co-morbidity but that does not detract from the causal relation to ME/CFS.
That is, assuming that there really are 'co-morbidities' for ME/CFS. That term is often linked to EDS and MCAS, which probably are not com-morbidities. I am not sure we have good evidence for any co-morbidities.
Reactive depression occurs in people with ME/CFS but it does with most chronic illness and probably has nothing to do with spontaneous depressive illness or what is often called Major Depressive Disorder, which is what genetic studies tend to look at. I would not call reactive depression a co-morbidity.
I was unaware of any link to anaemia and I doubt there is one. There might be dietary deficiencies due to food intolerance in ME/CFS but if they led to anaemia that would not be an independent 'co-morbidity' with a separate genetic risk base.
What I see as much more likely is that some common clinical problems like anaemia, insulin resistance or hyperlipidemia have turned up as increased in ME/CFS cohorts for reasons of bias in cohort selection (Hutan has discussed this in detail). They are co-morbidities with ME/CFS as defined by those cohorts that tend to get recruited for ME/CFS studies but that is not what we want ME/CFS to be defined as. We want to define it as a putative common pathogenic pathway and clinical presentation that will not necessarily be ideally represented by cohorts that get recruited.
I may have got this wrong but I think it would be worth trying to get it clear in our minds. I don't see variants from DecodeME as being separable off as 'causing co-morbidities' rather than ME/CFS.
Nightsong
Senior Member (Voting Rights)
Haven't been able to watch the video but I was under the impression that psychiatrists have largely abandoned the old idea of there being a difference between endogenous depression or melancholia and reactive/situational depression. Perhaps the couple of psychiatrists on S4ME could comment but I suspect both types would be considered MDD if the depressed mood is sufficiently severe and it is not categorisable as something else (bipolar, secondary to substance use, etc). MDD seems to be a broad syndromic category that includes a lot of reactive/precipitated-by cases.
And definitions of depression used in genetic studies may be broader still. From a depression GWAS (link) using the UK Biobank cohort:
And definitions of depression used in genetic studies may be broader still. From a depression GWAS (link) using the UK Biobank cohort:
Jonathan Edwards
Senior Member (Voting Rights)
I appreciate that psychiatrists may have realised that there is not going to be a clear dividing line but there is all the difference in the world between a psychotic depression with major biological features (which I have spent six months nursing) and chronic lowness associated with life circumstances. From what I have heard from the community ME/CFS is associated with despair and sometimes suicidal thoughts but in keeping with the context.
I agree that the genetic studies may range over a mixture of both but I doubt that many people with ME/CFS would be given a diagnosis of MDD?
My real point is that if we are considering co-morbidities that might have different and confounding genetic origins we probably do not have evidence for depression being likely to fall in that category, if indeed such a category makes sense. If gene variants sort with ME/CFS then there has to be some causal path from one to the other (with some caveats that do not impinge on this) irrespective of any forks in the causal chain leading to other problems, I think.
hotblack
Senior Member (Voting Rights)
Having experienced both they are very different. This is of course subjective and I can see why some externally would see similar elements and even diagnose both. But to me it says a lot about how poorly we understand both. And I wouldn’t be surprised to find some commonality or crossovers in neurological pathways at some point.
Jonathan Edwards
Senior Member (Voting Rights)
Which would be another argument for not considering depression as a co-morbidity but as a category with some (but limited) overlap both in mechanism and presentation. What I think may get lost in these genetic studies is the fact that diagnoses are not mutually exclusive categories. They overlap at all sorts of level of stratification and often identify different classes of concept, whether trigger-based, mechanistic or clinical or whatever.
ScoutB
Senior Member (Voting Rights)
Chris said a couple different things on co-morbid conditions. My sense is he would agree with what you wrote about anemia and depression.
This was the place anemia and depression were mentioned:
The anemia mention seemed like an offhand example to me. If there is indeed overlap it could just be down to people who get periods (and thus are at higher risk of anemia) being a larger chunk of ME/CFS patients than the general population?
Another related point was around 30 minutes. Chris said:
I believe some of the genes SequenceME finds might be based on a very small number of patients having mutations in that gene. i.e. The astrazeneca analysis for BTN2A1 is based on 8 patients vs 50 controls having mutations in that gene -- since there are more than 50/8x controls than patients, that made BTN2A1-mutations more common amid patients. (Correct me if I'm wrong on any of this). Chris mentioned that SequenceME will be about 3x the size of this Astrazeneca cohort, and they'd be looking for rare variants, so I imagine some genes could again be implicated based on just a handful of pwME having them?
The situation of interest might be if 8 people have ME/CFS, a specific gene somewhat broken, and *also* all 8 have a specific other disease as well. Perhaps they do ultimately have *the ME/CFS disease process* (maybe as a side effect of the second disease) but it does seem worth looking into.
Last edited:
Jonathan Edwards
Senior Member (Voting Rights)
Yes, that is helpful @ScoutB. I think I can see what Chris is meaning to say. And to make this intelligible to most people with ME/CFS it is difficult to say it any other way.
But to me 'fibromyalgia'' and 'IBS' are not separate diseases or co-morbidities. They describe recognised components of ME/CFS. This is where using 'ME/CFS' both to describe the syndrome we have identified and the putative 'core ME/CFS process' that we think justifies that syndrome category produces problems.
There will be a complex network of causal steps and each person's illness will involve a different overlapping domain of that network. Within their domain may lie the 'core process' for ME/CFS, fibromyalgia and/ or IBS, inasmuch as even these are truly separable.
I can see that some gene variants will lie on a segment of the network specifically related to pain, or to gut, rather than PEM, for instance. On the other hand I don't think we should expect any of these genes to lie exactly within the 'core process' in ME/CFS. All the genes may be at upstream or downstream places that just give a clue to something in between.
hotblack
Senior Member (Voting Rights)
This is where I think some of the PrecisionLife results were not presented as well as they could. They talked about heterogeneity and complexity a bit too much IMHO sometimes giving the impression of different groups or fragmenting the cohort. My interpretation or reading of the clusters of genes they’d identified was just what you say, they were describing involved areas or mechanisms but they were pointers to something somewhere else upstream or downstream.
It sometimes seems to happen with DecodeME too and I get there is some use. But the treasure map analogy seems @Simon M used here seems much more powerful.
DecodeME - A "Treasure Map" | DecodeME : The world's largest ME/CFS study | Institute of Genetics and Cancer
We have always thought that this study alone may not provide all the definitive answers we need. Instead, we see our project as mapping out what future research directions should be taken and what treatments could be considered. We think of it as drawing a treasure map for other researchers...
institute-genetics-cancer.ed.ac.uk