Post

1) 🧬 We did a new genetic analysis using DecodeME. We leveraged a database on chromatin openness in the human brain. More openness means DNA is accessible and potentially active in various cell types. The graph below shows the results for ME/CFS and various other diseases.

2) The data comes from a landmark paper by Li et al. "A comparative atlas of single-cell chromatin accessibility in the human brain." It uses single-cell ATAC sequencing to show where DNA is accessible in various cells in the brain.

3) DNA is wrapped around proteins called histones, forming a structure known as chromatin. ATAC sequencing allows scientists to measure where the chromatin is open and where gene expression is likely taking place for each disease.

4) This atlas is from only 3 donors but includes 1.1 million cells from multiple brain regions. The study grouped all these cells into 107 cell types that form 42 subclasses. The authors already tested 19 diseases, we added ME/CFS genetic data as well.

5) In the graph, each dot represents the results for one of the 107 cell types, while the shaded regions indicate the transition to another subclass. Like many brain traits, ME/CFS showed significant results for glutamatergic neurons such as projecting neurons from the cortex.

6) But there's also a second, weaker signal that points to medium spiny neurons (MSM) in the basal ganglia. This aligns with earlier analyses we did on RNA (gene expression) where MSM were among the top cell types associated with ME/CFS 👇

7) You can also clearly see the difference with Alzheimer or Multiple Sclerosis which have no peaks for neurons but for other cell types in the brain such as microglia or blood vessel cells. These do not show up for ME/CFS.

9) Caveat: we're amateurs and this s-LDSC method is complex so it's possible that we made errors. We were, however, able to largely replicate results for two other traits (schizophrenia and tiredness, correlation ~ 0.99) so hopefully that means DecodeME analysis is also reliable.