[Poll] Sex differences in whether mothers or fathers of people with ME/CFS also have ME/CFS-like traits

[Jonathan Edwards]

Yes, I meant it doesn't support a specific mutation(s) on the X chromosome explaining a large part of the heredity.

Agreed. I think a slightly different statement might also be important - that it doesn't support a specific mutation(s) on the X chromosome explaining a large part of the familial clustering. I think there is still an open question about how much of the familial clustering is strictly inherited.

 

[Trish]

I wouldn't dismiss X linked inheritance being a factor on the basis of 2 people with XY having a father with ME/CFS similar symptoms. While it's clear they could'nt have inherited it from their father's X chromosome, they could still have inherited it it via their mother, especially as most of our sample had neither parent with symptoms, so it's not a simple inheritance pattern.

 

[forestglip]

I wouldn't dismiss X linked inheritance being a factor on the basis of 2 people with XY having a father with ME/CFS similar symptoms.

I agree. It doesn't provide support for the idea but we'd need larger and more controlled studies to rule it out.

----

Unrelated thought: what if a GWAS did an analysis on the subset of only participants where the participants thought that there was a good chance that they inherited it from one or both parents? It's not going to be perfect, as some of it could still be due to shared environment. But wouldn't that help concentrate only on participants where there's more likely to be a genetic component, thus creating a stronger genetic signal?

I don't think the data is there to do this with DecodeME, but maybe an idea for the future.

 

[Jonathan Edwards]

Unrelated thought: what if a GWAS did an analysis on the subset of only participants where the participants thought that there was a good chance that they inherited it from one or both parents?

We have discussed analysing just multicase families, which seems to me pretty much what you are suggesting. Chris is aware of that but the numbers may not be suitable.

 

[forestglip]

We have discussed analysing just multicase families, which seems to me pretty much what you are suggesting. Chris is aware of that but the numbers may not be suitable.

Yes, I was thinking about the numbers, but my idea is a bit more informal, so could potentially get high sample size. It wouldn't be requiring diagnosis. It'd basically be a question like this poll, where 40% think their parent had something kind of like ME/CFS.

Yes, it's going to be very imperfect. But it still excludes all participants where the genetic signal is likely to be weaker, since their parent didn't show any signs of it.

So if 40% think their parent also had ME/CFS and only a quarter of that subset are correct, that still brings the proportion that inherited it from around 10% in the full GWAS to 25%.

Whether or not it outweighs the smaller sample size, I don't know. But maybe it's worth one more question during recruitment and one more GWAS subset.

 

[SugarSquared]

There are 60 responses to the poll, and they seem to be tapering off now, so taking a look:

I have nothing to say about what this means, but I love the table! It is very pretty and easy to understand.

 

[forestglip]

Another interesting thing about this data is how few people think both parents had something like ME/CFS. Only 1 out of 67 at this point. Compare that to 26/67 who think one parent had something like it.

I think that indicates that it's unlikely that the results are primarily based on people with ME/CFS just being primed to see ME/CFS in other people even when it's not there. If that was the case, I don't see why almost no one would select both parents.

Thus, I think that means that for the people that voted that they think they may have inherited it, it is fairly likely to be accurate for most. Of course, it's still possible they "inherited" it through the shared environment rather than the genes.