ME/CFS Science Blog
Senior Member (Voting Rights)
As I understand it, these are the controls already used in DecodeME.
Preprint Biological Insights from Genome-Wide Association Studies and Whole Genome Sequencing of [ME/CFS], 2026, Maccallini et al
- Thread starter forestglip
- Start date
Jonathan Edwards
Senior Member (Voting Rights)
Sure.
mariovitali
Senior Member (Voting Rights)
I wanted to summarise and better understand what the study by @paolo maccallini is suggesting. Before I begin, please have in mind that the reason for this post is because I want to get to the bottom of this. I do not mean to be ironic or dismissive. I want to know what is going on and for this I am reaching out to people who know better than me.
1) In the following post, @Jonathan Edwards suggests that there are no metabolic abnormalities in ME/CFS. @Hutan and @forestglip do you agree ? As a patient I would like to know : Are results from ALL metabolomic studies done on ME/CFS irrelevant? If yes, why this is so ? Tagging also @ME/CFS Science Blog. If the answer is no, which metabolomic studies actually tell us something of value ? The same question goes for immune-related studies. Do we agree that we have replicated findings of lipid abnormalities (e.g. Cholines) @Jonathan Edwards ?
2) If I understand correctly, because of the results of this study, synapses and eccentric medium spiny neurons could be the master regulator of disturbances we are seeing in ME/CFS patients ? In other words, all disturbances we are seeing could be downstream effects of impaired synaptic function? Is this what is implied?
3) Is the following true or false : Conventional MAGMA analysis can be limited because SNPs are typically assigned to genes by physical position or LD-based mapping. For complex regulatory loci, especially in brain, the causal gene may be hundreds of kilobases away and connected through chromatin looping. If true, what are the implications ?
4) Is it true that MAGMA/FUMA enrichment identify tissues and cell types whose expressed genes carry more association signal, but it does not prove the disease originates in those cells, nor does it identify causal variants or causal genes ?
5) @ME/CFS Science Blog do you have any concerns regarding this study's input data and analysis?
1) In the following post, @Jonathan Edwards suggests that there are no metabolic abnormalities in ME/CFS. @Hutan and @forestglip do you agree ? As a patient I would like to know : Are results from ALL metabolomic studies done on ME/CFS irrelevant? If yes, why this is so ? Tagging also @ME/CFS Science Blog. If the answer is no, which metabolomic studies actually tell us something of value ? The same question goes for immune-related studies. Do we agree that we have replicated findings of lipid abnormalities (e.g. Cholines) @Jonathan Edwards ?
2) If I understand correctly, because of the results of this study, synapses and eccentric medium spiny neurons could be the master regulator of disturbances we are seeing in ME/CFS patients ? In other words, all disturbances we are seeing could be downstream effects of impaired synaptic function? Is this what is implied?
3) Is the following true or false : Conventional MAGMA analysis can be limited because SNPs are typically assigned to genes by physical position or LD-based mapping. For complex regulatory loci, especially in brain, the causal gene may be hundreds of kilobases away and connected through chromatin looping. If true, what are the implications ?
4) Is it true that MAGMA/FUMA enrichment identify tissues and cell types whose expressed genes carry more association signal, but it does not prove the disease originates in those cells, nor does it identify causal variants or causal genes ?
5) @ME/CFS Science Blog do you have any concerns regarding this study's input data and analysis?
Jonathan Edwards
Senior Member (Voting Rights)
I don't think they are irrelevant but my understanding is that the vast majority of findings are within normal ranges and that the few exceptions have not been replicated. I suspect that people have been looking at metabolomics because in the past there has been a tendency to think of ME/CFS in terms of altered energy metabolism. People try to find things in their data that confirm their hunches. In the past ME/CFS research has been a backwater and the level of critique within the science community has been fairly low.
I think I asked if you knew of any metabolic abnormalities we should focus on @mariovitali ? I forget things, but cannot remember anything striking. Can you?
I think we are all very aware that the spiny neurons may be a blind alley but at least we have some pointers in that direction.
Answering the question you tagged me on, just my view.
I am sure that there are some existing reports of ME/CFS abnormalities of various sorts that are true, no doubt we have discussed some of them. The problem is that, right now, I don't think we know which ones they are. They are hidden among all the noise and the lack of replication. I wouldn't say metabolic studies are irrelevant. We need more, done really carefully.
I certainly am not ruling out issues with lipids. But, I can't recall any convincing evidence yet. Some that is a bit suggestive, I guess, among other studies that have too many problems to tell us anything reliable. I'd be happy to be pointed to something that people feel is convincing.
mariovitali
Senior Member (Voting Rights)
@Jonathan Edwards You did ask and I gave an example regarding replicated findings of disruption in Cholines and other phospholipids. Do you think that these should not be considered replicated results ? If yes, why this is so? @Hutan are lipids such as cholines considered replicated findings or not ?
If it is a matter of confidence then we should also be more confident regarding the study discussed in this thread hence my questions (3) and (4).
Thank you @Hutan. If there are some existing reports which we know are true and at the same time we don't know which these are , I am seeing two contradicting statements here. I do understand the lack of replication you mention and the better selection of cohorts and study design in general which you mentioned and I totally agree.
I mean, if after all these studies we do not have at least some directions.. this is frightening and deserves immediate communication to patients and patient groups.
Jonathan Edwards
Senior Member (Voting Rights)
Sorry, I must have missed that. Ther are so many threads.
I am not sure what this has to do with Paolo's suggestion. We are not running a theory competition. We are looking at everything that might provide clues.
I don't understand what is frightening. We have always known that we understand almost nothing about ME/CFS. Lots of negative findings help to narrow things down. Hutan suggests that some of the positive findings may turn out 'true' findings but the problem is that we don't know which ones that will be. We now have some important genetic leads and genetic data are inherently much more powerful than other studies because they imply causal relation, even if that may be complex and hard to pin down.
What were the choline and phospholipid findings?
Many of the SNP-gene assignments are likely not correct, making the analysis based on not perfectly accurate data. But many likely are. If it was mostly SNPs assigned to genes incorrectly, I would think we would not see any significant tissues, as opposed to seeing significant tissues that are totally wrong. I'm sure the specific pattern of tissues could change somewhat though with more accurate data.
Yes, the core idea of MAGMA is fairly easily interpretable: Imagine there are only 3 genes in the DNA. SNPs near gene A are very significant and gene A is highly expressed in the liver. SNPs near gene B are moderately significant and gene B is moderately expressed in the liver. SNPs near gene C are not at all significant and gene C is not expressed in the liver. MAGMA would thus highlight the liver as significant. If gene A, B, and C do not show this pattern of expression in, say, the kidney - maybe all genes equally expressed there - then the kidney would not be significant.
If the genes that are more important for a tissue's functioning (higher expression) are also near DNA variants that are more implicated in increasing risk for a disease (higher GWAS significance), and vice versa, then the implication is that the tissue (or another tissue with similar expression patterns) may be where altered functioning causes disease.
It doesn't prove the tissue is most important. Maybe the liver shows up as significant, but there's a small section of the pancreas that happens to have a very similar expression pattern to the liver, and would match the GWAS pattern even more closely, but it was not tested.
It's certainly not conclusive, but it's one more clue to the puzzle. Brain tissues being highly significant but tissues like the lung being not significant adds some confidence that the brain may be more involved in ME/CFS than the lung. As for the specific pattern of highly significant tissues or cell types, my thinking is that's it's probably early to say with much certainty that specifically the most significant tissue or cell type is the most important, and the pattern may or may not change with further data.
No, MAGMA doesn't identify causal variants. The mapping of significant variants to genes could be considered a crude method of identifying causal genes (the 13 "MAGMA genes" in DecodeME), but again, many of these mappings are likely wrong so any given specific gene should be taken with a grain of salt.
mariovitali
Senior Member (Voting Rights)
I don't understand your comment about running a theory competition and how it connects with Paolo's suggestion. I also do not understand the phrase "Hutan suggests that some of the positive findings may turn out 'true' findings but the problem is that we don't know which ones that will be". I did not ask about which ones will be true, but what we are seeing so far across studies.
Supposedly, one of the most replicated finding is disruption of certain lipids and from what I understand we are unsure for this as well ?
I am asking a very simple and specific question : Are low phospholipids considered a replicated finding?@Jonathan Edwards could you please give a concrete "yes" or "no" to this question ?
Some related studies :
1. https://www.nature.com/articles/s43856-024-00669-7
2. https://www.mdpi.com/1422-0067/23/14/7906
3. https://www.mdpi.com/2218-1989/10/1/34
Last edited:
Jonathan Edwards
Senior Member (Voting Rights)
OK, but I don't know enough about the nomenclature to be confident that this is replication of specific findings. The findings are presented differently in each abstract. It looks likely that there are difference in plasma lipid profiles. Can we be sure that these are not due to dietary shifts in people with ME/CFS? I think it likely that dietary patterns for people with ME/CFS are fairly atypical. The lipid profiles might also reflect some shared susceptibility in people with ME/CFS to unrelated illnesses involving insulin resistance or whatever.
These are all statistical findings showing a population shift but I doubt that any of them show the sort of difference from normal that might make us think we are looking at direct mediation of disease. I also find it hard to think of a disease mechanism that would put these lipid changes centre stage. The finding might be telling us about things going on in certain cell membranes, maybe nerve cells or lymphocytes. I find it easier to focus on specific pathways that might lead directly to the symptom pattern.
Last edited:
Mario, that first one is the Huang, Armstrong study of blood lipids in people with a label of ME/CFS in the UK Biobank. As I and others explained ad nauseam on the forum thread for it, that study unfortunately does not give us evidence of altered lipids in ME/CFS because the ME/CFS individuals are not compared with suitable controls. It seems highly likely that most, but perhaps not all, of the identified differences are the result of comparing the ME/CFS group, who have the co-morbidities typical of 50 and 60 year old people, with people who were selected to be particularly healthy in most respects.
I think we have wandered off the topic of this thread. Maybe we can pick up a discussion about the evidence for lipid dysregulation on another thread?
Patients and patient groups certainly do need to know that a lot of ME/CFS research, including biologically focussed research, doesn't take us forward. That's not frightening news though. Us understanding that is progress, because we can do things to make the research better, rather than simply applauding as things go round in circles.
I'm feeling more confident than ever that we have some good leads and good people (including the Armstrong team) working on the problem.
Last edited:
mariovitali
Senior Member (Voting Rights)
I'm going to sign off for my night, but set up a discussion thread on the evidence for disruption of phospholipids if you want @mariovitali and I'll catch up tomorrow. Or, I can make a thread tomorrow.
neophyte32
Senior Member (Voting Rights)
It's quite depressing to read your posts. I was so excited about the publication of Maccalini's paper, as it perfectly matches my symptoms. A false hope then? How is it possible that we've made virtually no progress (I'm not expecting reassurance in return, even though my health is critical) for so many years...? Ron Davis and OMF, for example, have had a lot of money, and I feel like nothing is moving in the right direction.
Jonathan Edwards
Senior Member (Voting Rights)
That isn't how some of the rest of us are feeling!!
neophyte32
Senior Member (Voting Rights)
Yes, Professor, you did correct me on my one somewhat general and peremptory statement. Sorry.
I know we're swimming in uncertainty and that you often remind us that the solution could arrive at any moment, tomorrow or in 10 years, okay, but unless you know things we don't in the inner workings of research, what's on the horizon in two or three years is a bit depressing (before daratumumab and SequencEM).
If we had dozens of Maccalinis working behind the scenes, that would be fantastic, but how can we move forward without dozens and dozens of brain autopsies, in-depth studies on the most severely affected patients that would require home visits for saliva and blood samples, etc.? I get the impression that this forum is the only one that understands the disease. I was so excited to read this preprint...
Jonathan Edwards
Senior Member (Voting Rights)
No way. I confidently predict that we will make significant progress before either of those is done.
It will be less than 10 years. I have actually often said it might have been last week and not known to us, too. I think that is a significant possibility.
We do. There are a dozen here for a start.
When we worked out enough about RA to use rituximab there was almost nothing that we had needed to do in terms of experiments. The difficult part was piecing together all the stuff lying about and filtering the useful stuff from the dross. That is what is going on now for ME/CFS. Only this week we have woken up to the fact that ME/CFS gives an unusual response to buspirone. I have already asked some colleagues if they would like to look at that again. The genetic stuff is falling in to place in a way that I suspect can only snowball.
Many of us think we might just have turned the biggest corner for twenty years.
You only need one lot of people and there is a big lot here. And there is Paolo too, and there are others who are getting the message.
Last edited by a moderator:
MrMagoo
Senior Member (Voting Rights)
What I understand as a lay person with little science knowledge is that where ME is concerned, there won’t be one sudden “we found the cause, it’s X” it’s going to be an process of looking further into areas, discovering stuff, even re-analysing previous clues, closer to completing a jigsaw puzzle piece by piece than finding a buried treasure.
That’s why science advances are exciting - because we found a corner piece and more of the edges. Once you have the edges you can re-evaluate those funny single-colour pieces that might fit in, maybe connect a patch of pieces.
But the puzzle is far from complete and the research is just a jumble of pieces, some might be useful, some might obviously be from a different jigsaw and not fit. Some piece that seemed odd and not likely to fit actually slots in perfectly when you find the surrounding ones and see the ideal space for it was there all along.
That’s why science advances are exciting - because we found a corner piece and more of the edges. Once you have the edges you can re-evaluate those funny single-colour pieces that might fit in, maybe connect a patch of pieces.
But the puzzle is far from complete and the research is just a jumble of pieces, some might be useful, some might obviously be from a different jigsaw and not fit. Some piece that seemed odd and not likely to fit actually slots in perfectly when you find the surrounding ones and see the ideal space for it was there all along.
Last edited:
V.R.T.
Senior Member (Voting Rights)
Obviously what you say here is hypothetical but do you mean it in the sense of a massive finding being discovered 'last week' but not written up/published by the lab yet, so we are all unaware?
That is an exciting idea!
Jonathan Edwards
Senior Member (Voting Rights)
It does not have to be a massive finding. It may be something that just happens to make sense of a lot of other stuff. All the key clues I had for RA had been published by someone else before. In one crucial case it turned out it had been previously published, by
us!