SequenceME genetic study - from Oxford Nanopore Technologies, the University of Edinburgh and Action for ME

[neophyte32]

This is all very encouraging!
In short, we have to wait 2 to 3 years before we get any initial answers about this disease.

This will coincide with the results of the Fluge and Mella trials. 2028/2029, a real challenge for those who are bedridden, but we have a goal.

 

[mariovitali]

At 42:50, Saddie Whitaker asks @Chris Ponting about the use of Large Language Models after having acknowledged their weaknesses.

@Chris Ponting then mentions about being in talks with"very high-powered machine learning organisations" . Will there be an update on which organisations these are? I wonder how larger the actual SequenceME data be and what kind of resources will be needed to apply machine learning methods to this kind of data.

 

[Andy]

Will there be an update on which organisations these are?

Should they prove to be fruitful talks then yes, I would imagine there would be.

I wonder how larger the actual SequenceME data be and what kind of resources will be needed to apply machine learning methods to this kind of data.

All I know is that over 1 petabyte of storage will be needed for the first 6k sequences.

 

[SugarSquared]

I’ve got a question about the results SequenceME will give us. Will it be able to pinpoint the exact genes are the SNP signals in DecodeME? Like, will it search for the treasure to find the specific culprit? I know that it will be able to look at many places that GWASes cannot, but I don’t know if it will be able to go over the same areas as the GWAS but be more specific. It would be cool to have less guesswork surrounding which gene is actually responsible.

 

[ScoutB]

I’ve got a question about the results SequenceME will give us.

I've been trying to read about this a bit and so far I've learned: it's complicated. In some cases SequenceME may 'agree' with DecodeME and point out specific genes in an area that DecodeME had only been able to loosely highlight. Like how the AstraZeneca data (which is like a poor man's SequenceME) agrees with DecodeME on BNT2A2/BNT2A1. In other cases SequenceME might point us to different genes than DecodeME did.

I read a few blog posts and articles on the 'genetic architecture' of neurological conditions recently and I think those examples might give us some idea of what SquenceME might find (if it points in a neurological direction as well). Researchers have been studying the genetics of these other conditions more intensely and so already know both the kind of information that we got from DecodeME (how common snp variants contribute to risk) as well as info we might get from SequenceME -- i.e. rare variants that singlehandedly raise risk a lot, including genetic changes that are more complex than just snp's.

Apparently these two kinds of genetic risk don't always overlap as much as you'd expect.

This paper is relevant I think:
Specificity, length, and luck: How genes are prioritized by rare and common variant association studies
It talks about how GWAS studies (like DecodeME) tend to identify variants that are involved in the condition but are also highly pleiotropic, meaning they also influence many other unrelated traits. On the other hand "rare variant burden" studies (which I believe SequenceME would count as) are said to find genes that are less pleiotropic and (maybe, hopefully) more closely related to trait biology.

Anyway, back to comparing with neurological conditions. I think I posted some quotes from this genetics blog already somewhere, but here are some more describing how our understanding of the genetics evolved:

In the early 2000’s, the prevailing view was that the genetics of conditions like autism or schizophrenia was complex (though it wasn’t at all obvious in what way). It was known that there were rare genetic conditions that could result in the symptoms of psychiatric disorders – like Fragile X syndrome or Rett syndrome in the case of autism, or velocardiofacial syndrome in the case of schizophrenia. However, these were deemed to be exceptional cases – the main pool of “idiopathic” cases (the ones with no known genetic or organic cause at the time) were considered by many to reflect the “real” conditions of autism or schizophrenia.

More recent studies are helping to identify these risk factors and elucidate the nature of their interactions. Some patients carry single de novo mutations that confer very high risk and are probably sufficient to explain the occurrence of disease by themselves. Other patients carrying mutations conveying lower statistical risk have been found to be more likely to also carry a “second hit” somewhere else in the genome and/or to have a higher polygenic burden of common risk variants. This is consistent with a unifying model whereby the polygenic risk acts as a genetic background modifier of the effects of rare, high-risk mutations. There is thus no need or reason to suggest that there are separate pools of patients accounting for these conditions – some entirely caused by rare mutations and some entirely caused by polygenic burden. Both factors are likely at play in most, if not all patients.

Basically, I'm wondering/thinking DecodeME and SequenceME could point at these two distinct kinds of genetic risk, if the genetics of ME/CFS ends up being similar to these other conditions.