A look at my whole genome sequencing results

Hoopoe

Hoopoe

Senior Member (Voting Rights)
I have access to my own raw data and am looking through it slowly.

One of the interesting findings so far is a series of mutations in the HLA-DQB1 gene, and one in the HLA-DRB1 gene which are all associated with increased risk for multiple sclerosis.

PS: also many variants in HLA-DQA1.

I also have one of the IDO2 variants reported by Phair as potentially damaging (which is very common in the general population, so it doesn't really mean anything), and one that is a missense variant not reported by him.
 
Last edited:
I also have a "likely pathogenic" variant expected to cause a nonfunctional protein in a gene involved in fatty acid oxidation, for which no population frequency data exists, and which has no entry on ClinVar. Sounds promising.

PS: but the associated disease has an autosomal-recessive inheritance pattern which means that this mutation might have no damaging effect, or only a small one. Maybe this is not the answer?
 
Last edited:
Small update on the fatty acid oxidation mutation.

After applying a healthy dose of skepticism and doing some research, I think this a genuine finding, with a real impact on enzyme activity (I found a paper with a patient that had the same kind of heterozygous mutation where enzyme activity was measured). Whether it is actually behind certain symptoms in my family is a bit tricky to determine though. I really doubt this alone could explain my ME/CFS.
 
Last edited:
Did you get any analysis or panels that group relevant genes together into reports?

Things like Immunology, Neurology, Muscular Dystrophy, Connective Tissue Disorders, Periodic Fevers, Cardiology, Gastroenterology, Periodic Paralysis, Opthalmology, Mitochondrial Panel etc etc

I bought all these reports when on sale at reasonable cost. Some have been helpful, some not very helpful.

Some things I found on 23andme were validated by the WGS.

Worth looking into your heritage eg I have a lot of Irish heritage so am looking at Irish heritage diseases.

I also have some Sardinian heritage so looking at their heritage diseases.

Only got the heritage through 23andme plus records from own family history.
 
Small update:

The fatty acid oxidation mutation found is in the gene corresponding to the disease "medium chain acyl-CoA dehydrogenase deficiency". As mentioned this is an autosomal recessive disease which would not be expected to manifest when only one of the two inherited genes is affected. Some people believe in the idea "synergistic heterozygosity" which says that a person can have inherited diseases if multiple genes affecting the same pathway have mutations (that individually would not be able to cause disease). The symptoms of this illness are poor fasting tolerance, low energy and episodes of hypoglycemia which fits very well. What does not fit is that this disease also causes episodes of life threatening metabolic decompensation, but that might be a question of severity. If I have this, it would also be the "late onset" form which is recognized to exist and may have a slightly different symptom profile. I also have a mutation in a similar gene that is involved in fatty acid oxidation of longer fatty acids but it's less clear if this mutation is pathogenic. The other mutation is almost certainly pathogenic.

In this study, some patients with only a heterozygous mutation had residual enzyme activity that is in the range where it is of some concern (according to what I've read, it's clearly not in the range of severe illness). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3444485/
 
Last edited:
Did you know that we all have a huge amount of genetic mutations? In some of my genes I have hundreds of mutations, in others just five or so. The difficulty is figuring out what these mutations do, and which ones are relevant to the problem we're trying to solve. We don't seem to know what the vast majority of mutations actually do. I'm also constantly seeing mutations which are not listed in databases. That might suggest they're very rare.

I'm mainly interested in finding out what is causing my poor fasting tolerance / episodes of hypoglycemia. This is very likely a genetic problem.

Some other interesting mutations:

My SC25A5 gene has three mutations that seems like they might lead to a nonfunctional protein. This protein exchanges ATP and ADP between cytosol and mitochondria. It's one of three proteins that have this function. I haven't figured out yet how to tell in this kind of situation if both copies of the gene are affected or just one. The allele frequency of these mutations is relatively high so these are maybe not particularly impactful, somehow.

SLC44A4 also looks interesting because it has a mutation that might affect phospholipid synthesis (relevant for ME/CFS according to recent research by Lipkin et al) and synthesis of acetylcholin.

SLC25A37 is another mitochondrial protein. There is a missense mutation. The function of this protein is iron import into mitochondria. The mutation is very rare and in and a highly conserved region. There is no data on what this mutation might be doing. It is however predicted to be benign by most prediction tools.
 
Last edited:
I spoke to a neurologist who agreed with everything I said about the possibility of "medium chain acyl-CoA dehydrogenase deficiency". He believes it would show up on my muscle biopsy. I'm waiting for the results of that.

PS: and based on what I've read, this disease would not be able to explain significant functional impairment in daily life, unless the person had suffered from severe hypoglycemia at some point that caused irreparable damage. This would add to my diagnoses instead of replacing ME/CFS.
 
Last edited:
Another interesting mutation in the PRKAG3 gene, although difficult to interpret and not clearly pathogenic.

Wikipedia said:
The protein encoded by this gene is a regulatory subunit of the AMP-activated protein kinase (AMPK).

AMPK is an important energy-sensing enzyme that monitors cellular energy status.

Studies of the pig counterpart suggest that this subunit may play a key role in the regulation of energy metabolism in skeletal muscle.
Click to expand...
 
I found something that might be part of the genetics of ME/CFS or associated illnesses. The chance that this is important is probably low but it's still encouraging. We should be studying more families where people have ME/CFS and associated illnesses like POTS or orthostatic hypotension. I'm in contact with a researcher about this. The associated mutations are there, we just have to find them.
 
Last edited:
Some genes have a large number of variants. For example my HLA-DQA1 gene has 1244 :eek:. Here the software highlights 17 of them as being of possible interest. None of these are probably doing anything significant. The HLA genes seem to be the among worst in terms of sheer number of mutations.

HLA-DQA1.JPG

Other genes have very few or only a single variant, like for example my IGSF8 gene. The software flagged this because there's a frameshift mutation which is generally damaging. However not much is known about this gene.

I also had a blood test to rule out the possibility of antitrypsin deficiency as I had two potentially high impact mutations in the relevant gene, but my blood levels were high enough to avoid problems.
 
Last edited:
deleder2k said:
How did this turn out?
Click to expand...

While there were many potentially pathogenic variants in my data, there was no clear evidence of a genetic disease.

I did find some variants that probably constitute risk factors for various problems. Knowing that I have elevated risk for certain conditions is not particularly helpful.

Zhang et al reported the ACE gene as being associated with ME/CFS, which aligns with what I suspected (I have a ACE mutation). Another gene that came up in my data was closely related to circadian rythm which has come up in the Precision Life data. I'm not sure how my ACADM mutation affects me, it might turn out to be relevant. There were some other things that might matter... but how does one confirm this, and benefit from the knowledge in concrete ways? That is where things get difficult.

There is probably a lot of interesting information that could be gained from WGS data but it is time consuming and difficult to work with. I'm not sure that I ever managed to properly analyze my data the way an expert would be able to. The problem the expert has is that they tend to struggle understanding the clinical picture and linking it to a disease unless there are some tell-tale symptoms and signs consistent with a known genetic disease.

Was it useful? Yes, to rule out some possibilities. Was it interesting? Yes, I learned a lot. It's an interesting pastime. Did it help me? No.
 
Last edited:
Hoopoe said:
I have access to my own raw data and am looking through it slowly.

One of the interesting findings so far is a series of mutations in the HLA-DQB1 gene, and one in the HLA-DRB1 gene which are all associated with increased risk for multiple sclerosis.

PS: also many variants in HLA-DQA1.

I also have one of the IDO2 variants reported by Phair as potentially damaging (which is very common in the general population, so it doesn't really mean anything), and one that is a missense variant not reported by him.
Click to expand...
Its interesting to see this after the DecodeME HLA-DQ link. I think JE said yesterday that F&M and maybe Wessely (bleurgh) also found a DQ link before that.
 
V.R.T. said:
Its interesting to see this after the DecodeME HLA-DQ link. I think JE said yesterday that F&M and maybe Wessely (bleurgh) also found a DQ link before that.
Click to expand...
It probably doesn't mean anything. HLA-DQA1 is a highly variable region. Everyone would have pothentially pathogenic variants according to variant effect prediction tools. These tools are made for analyzing "normal" genes. Some of them could affect the risk of ME/CFS but there is no way of knowing.
 
Last edited:
I looked through my raw data on 23nMe and could not find any ClinVar reported polymorphism relating to the list from Decode. I am a monozyg for HLA DRB1 1501 which is a hit for chronic, esp pesticide and organic dust induced sarcoid and for fatigue states but found nothing else. I have a shed load of "raisers of index of suspicion for Lyme". Does the failure to find anything on the decode ME genes indicate lack of/low ME propensity of will any links with ME just not yet be noted in clinvar or is my approach wrongheaded?
 
Hoopoe said:
While there were many potentially pathogenic variants in my data, there was no clear evidence of a genetic disease.

I did find some variants that probably constitute risk factors for various problems. Knowing that I have elevated risk for certain conditions is not particularly helpful.

Zhang et al reported the ACE gene as being associated with ME/CFS, which aligns with what I suspected (I have a ACE mutation). Another gene that came up in my data was closely related to circadian rythm which has come up in the Precision Life data. I'm not sure how my ACADM mutation affects me, it might turn out to be relevant. There were some other things that might matter... but how does one confirm this, and benefit from the knowledge in concrete ways? That is where things get difficult.

There is probably a lot of interesting information that could be gained from WGS data but it is time consuming and difficult to work with. I'm not sure that I ever managed to properly analyze my data the way an expert would be able to. The problem the expert has is that they tend to struggle understanding the clinical picture and linking it to a disease unless there are some tell-tale symptoms and signs consistent with a known genetic disease.

Was it useful? Yes, to rule out some possibilities. Was it interesting? Yes, I learned a lot. It's an interesting pastime. Did it help me? No.
Click to expand...

I would use Franklin Genoox. The University Hospital here use it. It is free. Upload your VCF. Add phenotypes (i.e muscle pain, fatigue and so on). You also need to adjust the allelle frequency (you don’t want to se common variants), then go to prediction and then you can look for variants with a REVEL score above 0.5 or 0.7. ChatGPT can help you with it all. Copy paste your results to ChatGPT and ask for a deep analysis of the variants, and whether they can explain your situatuon.
 
You must log in or register to reply here.
Back
Top Bottom