Examination of Single Nucleotide Polymorphisms in Acetylcholine Receptors in Chronic Fatigue Syndrome Patients 2015 Marshall-Gradisnik et al

Abstract

Objective
Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is a disorder characterized by debilitating fatigue accompanied by pain and impairments in memory, cognition, and concentration. Acetylcholine (ACh) has a plethora of roles in neuronal and neuromuscular transmission. There are two types of ACh receptors, muscarinic and nicotinic, comprising 17 different subunits of the nicotinic ACh receptor (nAChR) and five different subtypes of the muscarinic receptor (mAChR) that have been identified in humans. The purpose of this study was to determine the role of ACh receptor (nAChRs and mAChRs) single nucleotide polymorphisms (SNPs) in CFS/ME patients.

Methods
One-hundred and fifteen CFS/ME patients (age = 48.68 ± 1.06 years) and 90 nonfatigued controls (age = 46.48 ± 1.22 years) participated in this study, where CFS/ME patients were defined according to the 1994 Center for Disease Prevention and Control (CDC) criteria. A total of 464 SNPs for nine mammalian ACh receptor genes (M1, M2, M3, M4, M5, alpha 2, 5, 7, and 10) were examined via the Agena Biosciences iPLEX Gold assay. Statistical analysis was performed using the PLINK analysis software.

Results
Seventeen SNPs were significantly associated with CFS/ME patients compared with the controls. Nine of these SNPs were associated with mAChRM3 (rs4463655; P = 0.00281, rs589962; P = 0.00348, rs1072320; P = 0.00371, rs7543259; P = 0.00513, rs6661621; P = 0.00536 rs7520974; P = 0.0167, rs726169; P = 0.02349, rsrs6669810; P = 0.02361, rsrs6429157; P = 0.0375), while the remainder were associated with nAChR alpha 10 (rs2672211; P = 0.0107, rs2672214; P = 0.0108, rs2741868; P = 0.01185, rs2741870; P = 0.01281, rs2741862; P = 0.03043), alpha 5 (rs951266; P = 0.01153; rs7180002, P = 0.03682), and alpha 2 (rs2565048; P = 0.01403).

Conclusion
The data from this pilot study suggest an association between ACh receptors, predominantly M3 and CFS. ACh receptor SNPs may contribute to the pathomechanism of CFS/ME.

Open access, https://journals.sagepub.com/doi/full/10.4137/III.S25105

 

They screened, 464 SNP’s, and the study make no mention of correcting for the many comparisons (they said they set the value at p less than 0.05: they would be talking about q values if they had corrected for multiple comparisons).

This looks like a study of false positives (this team has made this error before).

 

But it would be very easy to run the stats on the UK Biobank sample with a Bonferroni of only 464 and a direction for the signal for each SNP?

 

It would take a bit of time, it wouldn’t be too difficult for someone with the right skills.

Simply applying Bonferroni to this data, the adjusted value would be 0.00013 (they used two cohorts and only 396 SNPs were common both). The closest value here is 0.0028, more than 20 fold higher, which seems like a clear null result,

 

One of the UK Biobank GWAS studies has a nice interactive website (http://geneatlas.roslin.ed.ac.uk/).

I thought I'd check if at least any of these variants had at least a halfway lowish p value, but none of their unadjusted p values get close to satisfying alpha = 0.05 even before multiple testing correction.

 

--- yip even I recall that - so why didn't they!

 

Cool.