You’ve read the first three posts in this series. A post-GWAS cell-type analysis implicates medium spiny neurons. The striatum is the bottleneck. Dopamine drugs are suddenly rational. It is a clean, satisfying story — and like all clean, satisfying stories in biology, it has a problem.
The signal the genetics points to is not an ME/CFS fingerprint. Brain-neuron enrichment shows up in many brain disorders.
Shared in the brain, different in the cell
ME/CFS is not the only condition whose genetic risk enriches in brain neurons. But “brain neurons” is a broad category, and the specific cell types differ between conditions. Applying the same atlas-based cell-type pipeline across traits gives:
So it is not quite right to say these diseases share the same cell type. What they share is that their genetic risk is expressed in the brain’s neurons rather than in immune cells. The finer cell-type resolution differs — and on the same pipeline, ME/CFS and alcohol use disorder are the two that converge on MSNs, while schizophrenia and depression converge on cortical neurons.
Brain vulnerability is broad. ME/CFS, schizophrenia, depression, and alcohol use disorder all affect brain circuits — but through different cell types and different molecules. The shared feature is the organ, not the fingerprint.
But the diseases are plainly not the same. A schizophrenia patient experiences hallucinations and thought disorder. A depressed patient experiences anhedonia and worthlessness. An ME/CFS patient experiences post-exertional malaise and orthostatic intolerance. Different cell types, different phenotypes — a signal about what we are measuring and what we are missing.
Why shared enrichment doesn’t mean shared mechanism
GWAS cell-type enrichment tells you where risk variants are expressed, not what those variants do. An SNP enriched in a neuron type could affect any of the hundreds of genes expressed in those cells. The specific genes, the direction of effect, the developmental timing, and the interaction with other cell types all differ between diseases.
The schizophrenia GWAS implicates synaptic pruning genes (C4, complement pathway), consistent with excessive synapse elimination during adolescence. The depression GWAS implicates genes involved in stress-hormone signalling and neurotrophic factor regulation. The alcohol use disorder GWAS implicates opioid and dopamine receptor genes (Trubetskoy et al. 2022; Wray et al. 2018; Zhou et al. 2020).
For ME/CFS, we do not yet know which specific genes within MSNs are affected — if MSNs are indeed where the signal concentrates. The cell-type result is a preprint, and it does not yet tell us whether the mechanism is:
- D1/D2 pathway imbalance — a shift in the “go”/“no-go” signal ratio that makes action initiation neurologically expensive
- Metabolic insufficiency — MSNs are GABAergic neurons that fire tonically at high rates; if mitochondrial function is globally impaired, MSNs fail first
- Dopamine receptor desensitisation — chronic low-grade inflammation downregulating dopamine signalling independent of genetic variation
- Cortico-striatal synapse dysfunction — impaired input integration from the frontal cortex to the striatum
These are not equivalent. A D1/D2 imbalance calls for dopamine agonists. Metabolic insufficiency calls for mitochondrial support. Receptor desensitisation calls for anti-inflammatories. Synapse dysfunction calls for something else entirely. Four different mechanisms, four different treatments. The shared brain enrichment does not weaken the finding — it sharpens the question: which mechanism within these neurons is specific to ME/CFS?
What the shared finding cannot do
The shared brain-neuron enrichment does not support several claims that a naïve reading might suggest:
It does not make ME/CFS a psychiatric disease. Schizophrenia and depression are also brain diseases with genetic architecture expressed in neurons — and they are also not “just in your head” in the dismissive sense. Brain enrichment shows these conditions share a biological organ, not a cause or a treatment. The brain is an organ. Diseases of that organ are neurological.
It does not justify prescribing antipsychotics or antidepressants for ME/CFS. Enrichment in a cell type says nothing about whether the drugs developed for one condition work for another. Most antipsychotics block D2 receptors — the opposite of what the dopamine-deficit model of ME/CFS would predict. Cell-type overlap must not be confused with pharmacological convergence.
It does not weaken the genetics argument for neurological primacy. The finding that brain-neuron enrichment is shared with other conditions does not weaken the finding that immune cell enrichment is absent. The genetics still say: the disease’s constitutional risk is expressed in neurons, not immune cells. The distinction between “it’s in the brain” and “it’s in the immune system” remains the critical one.
The next question the field must answer
If this series has argued for three conclusions — (1) genetics points to the neurological locus, (2) the striatum is a candidate bottleneck, (3) dopamine drugs become a rational thing to test — then the fourth post adds the necessary qualifier: (4) brain-neuron enrichment is shared with other brain disorders, and the specific dysfunction within MSNs (if that is the right cell type) that produces ME/CFS remains unknown.
This is not a weakness. It is the next research question, and it is a precise one: what is the molecular signature of MSN dysfunction in ME/CFS that distinguishes it from the pathology of other brain disorders? Answer that, and you have a drug target — not a shared circuit, but a disease-specific lesion within it.
Until then, every dopamine drug for ME/CFS is targeting a plausible circuit for a mechanism we cannot yet specify. Rational pharmacology, but at lower resolution than we need.
Certainty estimate
| ME/CFS genetic risk enriches in brain neurons, not immune cells |
High — GWAS tissue-level finding; immune null |
| Medium spiny neurons are the enriched cell type in ME/CFS |
Low-to-moderate — from a post-GWAS cell-type analysis (preprint), atlas-dependent; not in the GWAS itself |
| Different brain disorders enrich different cell types |
High — a common atlas pipeline puts schizophrenia in cortical layer 2/3, depression in layer 5/6, and AUD in MSNs (Duncan et al. 2025) |
| Shared brain enrichment does not imply shared mechanism |
High — different diseases involve different genes, pathways, and molecular pathologies |
| The specific molecular lesion in ME/CFS striatal neurons is unknown |
Very high — this is a statement of what has not been done, not what the evidence shows |
| Transcriptomic or PET profiling could distinguish ME/CFS striatal pathology from other conditions |
Moderate — the strategy has worked in psychiatry; has not been applied to ME/CFS |
| D1/D2 imbalance is the most likely mechanism |
Low — one of several candidates; no direct evidence in ME/CFS specifically |
Post 4 of 4 in the DecodeME GWAS series. Prev: The Genetics Settled It · The Striatum Is the Bottleneck · Parkinson’s Drugs for ME/CFS?
References
DecodeME Consortium, Chris P Ponting, et al. 2025.
“Initial Findings from the DecodeME Genome-Wide Association Study of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome.” medRxiv.
https://doi.org/10.1101/2025.08.06.25333109v1.
Duncan, Laramie E, Tony Li, Madeleine Salem, Will Li, Leili Mortazavi, Hazal Senturk, Narges Shahverdizadeh, et al. 2025.
“Mapping the Cellular Etiology of Schizophrenia and Complex Brain Phenotypes.” Nature Neuroscience 28 (2): 248–58.
https://doi.org/10.1038/s41593-024-01834-w.
Maccallini, P. 2026.
“Biological Insights from Genome-Wide Association Studies and Whole Genome Sequencing of Myalgic Encephalomyelitis/ Chronic Fatigue Syndrome.” Research Square [Preprint], June.
https://doi.org/10.21203/rs.3.rs-9702020/v1.
Trubetskoy, Vassily, Antonio F. Pardiñas, Ting Qi, et al. 2022.
“Mapping Genomic Loci Implicates Genes and Synaptic Biology in Schizophrenia.” Nature 604 (7906): 502–8.
https://doi.org/10.1038/s41586-022-04434-5.
Wray, Naomi R., Stephan Ripke, Manuel Mattheisen, et al. 2018.
“Genome-Wide Association Analyses Identify 44 Risk Variants and Refine the Genetic Architecture of Major Depression.” Nature Genetics 50 (5): 668–81.
https://doi.org/10.1038/s41588-018-0090-3.
Zhou, Hang, Julia M. Sealock, Sandra Sanchez-Roige, et al. 2020.
“Genome-Wide Meta-Analysis of Problematic Alcohol Use in 435,563 Individuals Yields Insights into Biology and Relationships with Other Traits.” Nature Neuroscience 23 (7): 809–18.
https://doi.org/10.1038/s41593-020-0643-5.