Genetic data
Let’s now move on to some actual data. The best evidence that we have is from genetic studies because of large sample sizes that control for population differences and that are not confounded by other factors such as diet, behavior, or the illness itself.
If we look at all the differences found in DecodeME (not just the 8 hits), check which genes they affect and where these are expressed, then the answer is the brain. The results of the MAGMA analysis are shown below.
View attachment 27905
Forestglip (
@forestglip ) has done a
preliminary analysis that looks at more specific cell types and here neurons came out on top. A
gene set analysis pointed to the synaptic membrane.
View attachment 27906
View attachment 27907
If we look at the hits that reach statistical significance, the closest genes often point to the same location.
The best example is perhaps the hit on chromosome 17 where the gene CA10 is the only candidate, and it is clearly linked to
neurons and synapses.
CA10
UNC13C seems the closest to the hits on chromosome 15. It’s gene card reads:
POU3F2 is the closest gene to the hit on chromosome 6q
PEBP1 seems like the second closest to the hit on chromosome 12, next to TAOK3, which seems very stretched out.
For the hits on chromosomes 1 and 20, there are so many potential genes in the location that it's harder to guess which one might be relevant. The most plausible hits on chromosome 13 (OLFM4) and chromosome 6p (butyrophilin3 and -2 homologues (BTN2A2, etc.) point to the immune system.
I don’t think this can be a coincidence. The DecodeME study focuses a lot on colocalization and gene expression to find causal genes, but there are some caveats with this approach, as explained here. I put more focus on the gene that is closest to the SNP hit, as previous research found this is more likely to be a causal gene than those that are further away.
Neurons and their synapses were also highlighted by the Zhang et al. preprint by the group of Michael Synder. They took a different approach, focusing on rare variants that are likely to lead to a loss of function using data on protein interactions and a neural network. They reported: “As highlighted in our network analysis, ME/CFS genes participate in biological pathways associated with synaptic function.” This was not expected, and they were the first to report this. They found several synapse-related genes, such as SYNGAP1 or genes like NLGN1, GRM157, DLGAP1 that code for proteins that regulate synaptic function.