Genetics: locus 20: 33,363,039 - genes CDK5RAP1, SNTA1, etc.

[ME/CFS Science Blog]

Another locus that was just below the significance threshold (p = 5.4 × 10⁻⁷) in DecodeME but unfortunately, it's hard to determine which gene it points to.

There are lots of genes in this region and the credible set is very spread out. Many of the SNPs are inside SNTA1 and CBFA2T2 so they are candidates to consider.


In neurons there were also some Hi-C data pointing to other genes such as NECAB3, C20orf144, ACTL10, CBFA2T2, etc.

 

[ME/CFS Science Blog]

The eqtl data didn't show much but the phewas showed that many traits share this signal including medication use, protein levels and mean corpuscular volume.

(Many of the studies and traits below probably overlap in dataset - I didn't take the effort to filter them out).

Trait	PP.H4	lead_p	study_ID	Source
Medication use (HMG CoA reductase inhibitors)	0.99	2.2e-09	ebi-a-GCST90018989	OpenGWAS
Hyperplasia of prostate	0.99	7.2e-08	finn-b-N14_PROSTHYPERPLA	OpenGWAS
Total protein	0.99	2.9e-09	ukb-d-30860_raw	OpenGWAS
red cell diameter width (RDW, minimum, inv-norm transformed)	0.99	2.0e-13	GCST90476365	Open Targets
Mean corpuscular volume	0.99	1.6e-10	ebi-a-GCST90018966	OpenGWAS
Mean corpuscular volume (UKB data field 30040)	0.99	9.1e-14	GCST90474481	Open Targets
Total protein	0.99	3.9e-09	ukb-d-30860_irnt	OpenGWAS
Serum total protein levels	0.98	3.1e-10	GCST90319650	Open Targets
Total protein levels	0.98	1.3e-11	GCST90691582	Open Targets
Mean corpuscular volume	0.98	7.5e-14	ebi-a-GCST90002334	OpenGWAS
Total protein levels (UKB data field 30860)	0.98	5.7e-09	GCST90310183	Open Targets
Pan cancer	0.98	1.1e-09	ieu-b-4964	OpenGWAS
Mean corpuscular haemoglobin levels (UKB data field 30050)	0.98	3.6e-15	GCST90474486	Open Targets
Mean corpuscular volume (UKB data field 30040)	0.97	1.7e-12	ebi-a-GCST90013979	OpenGWAS
Mean corpuscular volume	0.97	9.1e-13	ebi-a-GCST90002392	OpenGWAS
Mean corpuscular hemoglobin	0.97	3.2e-10	ebi-a-GCST90018964	OpenGWAS
Diseases of male genital organs	0.97	7.9e-07	finn-b-N14_MALEGEN	OpenGWAS
Red blood cell (erythrocyte) count (UKB data field 30010)	0.97	3.8e-09	GCST90691810	Open Targets
Mean corpuscular hemoglobin	0.96	8.9e-18	ebi-a-GCST90028995	OpenGWAS
Mean corpuscular hemoglobin	0.96	3.0e-13	ebi-a-GCST90002322	OpenGWAS

 

[ME/CFS Science Blog]

So probably not a region to read to much into, but posting it anyway in case the situation changes in the future when we have more data.

SNTA1 encodes alpha-1-syntrophin, a cytoplasmic peripheral membrane scaffold protein that organizes signaling complexes in muscle and cardiac cells. It is a member of the syntrophin family and the most common syntrophin isoform found in cardiac tissues. In these settings, SNTA1 contributes to the dystrophin-associated protein complex and is positioned at the neuromuscular junction, sarcolemma, anchoring junction, and cytoskeleton.

The protein’s PDZ domain binds the C-terminus of SCN5A, linking it to the cardiac sodium channel Nav1.5, and it also associates cardiac sodium channels with the nitric oxide synthase-PMCA4b complex in cardiomyocytes. Through these interactions, SNTA1 participates in regulation of sodium ion transmembrane transport, ventricular cardiac muscle cell membrane repolarization, heart rate, synaptic signaling, and muscle contraction. Expression is high in the brain, respiratory system, and proximal digestive tract, with strong abundance in cardiovascular system and muscle tissue.

Pathogenic variants in SNTA1 are reported in long QT syndrome and atrial fibrillation, and the gene is a susceptibility locus for long QT syndrome 1, 4, and 12. These associations place SNTA1 in disorders affecting cardiac rhythm and sudden arrhythmic death risk, and it has also been linked to sudden infant death syndrome.

SNTA1 Gene - GeneCards

 

[Jonathan Edwards]

So probably not a region to read to much into

Thanks goodness, my input channel is getting a bit full!!

 

[ME/CFS Science Blog]

Thanks goodness, my input channel is getting a bit full!!

Sorry about that! I'm posting these systematically for all regions that had a signal < 10^-6 in DecodeME, not just the ones I find interesting or valuable. So feel free to ignore these.

I plan to provide an overview afterwards of the genes that we have some confidence in as being linked to ME/CFS. This one probably won't make the cut because there are far too many candidate genes. The multiple traits that colocalize so strongly, also makes me think that this might be a locus driven by selection bias.

 

[ME/CFS Science Blog]

The main goal is to see if there any patterns in the gene's that come up if we make a longer list of (potential) ME/CFS genes. I'm use the FLAMES algorithm to help guide me search and select the ones that seem most reliable. For this locus it also couldn't make a selection.

 

[SugarSquared]

I plan to provide an overview afterwards of the genes that we have some confidence in as being linked to ME/CFS. This one probably won't make the cut because there are far too many candidate genes. The multiple traits that colocalize so strongly, also makes me think that this might be a locus driven by selection bias.

I’m really excited for this overview! You’ve been doing lots of impressive work.