The 12th Invest in ME Research Conference June, 2017, Part 2
MEMum presents the second article in a series of three about the recent 12th Invest In ME International Conference (IIMEC12) in London.
Discuss the article on the Forums.

Mitochondrial disease confirmed!

Discussion in 'Genetic Testing and SNPs' started by pattismith, Feb 12, 2018.

  1. pattismith

    pattismith Senior Member

    Messages:
    1,479
    Likes:
    2,235
  2. BeautifulDay

    BeautifulDay Senior Member

    Messages:
    310
    Likes:
    884
    For those following along, you go to Mitomaster for the SNV query.
    https://www.mitomap.org/mitomaster/index_snvs.cgi

    For Mitochondrial Mutations, enter the Ref Seq, the chromosome position, and the Var Seq.

    On @pattismith above from Enlis, it looked like M:3,377 A G
    I entered in Mitomaster A3377G in Step 2. Then on the next page click on Query
     
  3. pattismith

    pattismith Senior Member

    Messages:
    1,479
    Likes:
    2,235
    thank you , I finally managed it...

    I'am very suspicious now, do you think 23andme made some mistakes? Maybe I should ask them if they found frequently these 4 mutations in test, because it would indicate a possible error...
     
  4. BeautifulDay

    BeautifulDay Senior Member

    Messages:
    310
    Likes:
    884
    WIth 23andme it is possible. I've had 23andme call some variants that were tested by medical labs and were confirmed. I had one called by 23andme which was tested by a medical lab and was found to be a mis-call by 23andme.

    The way it works with 23andme is genotyping. Genotyping is not as good as WGS and WES. 23andme does have an error rate. Yet, for many people it's the most cost effective approach that fits within their budget. I found out a lot with 23andme (several of my mutations were medically confirmed), so it's a matter of having the real troublesome ones confirmed.

    I'll guess that 23andme's error rate is 1/2 percent (.5% error rate). (I believe it's somewhere close to that 99.5% correct rate)). It doesn't sound like a high error rate. I'll guess that 23andme looks at 500,000 SNPs on a chip. Using that error rate, there would be about 2500 errors per person tested. Most might be in those hidden internal numbers.

    Then you have a program that looks for rare mutations. Surely, some of those errors will pop up in that list. The rate will be higher than the original rate (here I estimated .5% error rate) due to the way we filter for rare variants. Errors will rise when filtering this way.
     
    pattismith likes this.
  5. Wishful

    Wishful Senior Member

    Messages:
    734
    Likes:
    1,077
    I don't remember where I read that cerebral mDNA differed from the rest of the body, but it seemed like a plausible source at the time, which might have been ten years ago. I'm not feeling up to searching for it today (PEM).
     
  6. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    so here are my results for mito variations:
    1)MT-CYB rs3088309 ....but it has been merged into rs527236209 Likely pathogenic. Ovarian cancer.
    2)MT ND5 rs28359178 . leber's optic atrophy.NADH dehydrogenase (complex 1)
     
    Last edited: Feb 15, 2018
    echobravo and pattismith like this.
  7. pattismith

    pattismith Senior Member

    Messages:
    1,479
    Likes:
    2,235
    are you sure 23andme tested your MT-ND5 rs28359178 ?

    I went in my raw datas for the MT-ND5 gene and couldn't find this rs number, so I wonder...

    Here on the picture, the rs28359177 is at the position 13590

    Would you go in your raw data and check which position is for the rs28359178? The best would be to do a copy of the screen and post it.

    upload_2018-2-15_16-36-12.png
     
    Last edited: Feb 15, 2018
  8. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    Genes Marker (SNP) Genomic Position Variants Your Genotype
    MT-ND5 rs28359178 13708 A or G A
     
  9. pattismith

    pattismith Senior Member

    Messages:
    1,479
    Likes:
    2,235
    You have two missense variants in mt DNA, and some mito-diseases have already been associated with them!
    Did you already consulted a mito specialist to see if they could be involved in your CFS/ME?
     
  10. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    No ....I found them last nite & not sure who I would reach out to .....plus are you sure they have been associated w/disease? Need to check on that.
     
  11. pattismith

    pattismith Senior Member

    Messages:
    1,479
    Likes:
    2,235
    the association with Leber Optic Atrophy remain uncertain, so nothing sure .
    Many carriers of mutations known to be associated with Leber optic atrophy will not experience any symptoms, whereas some will have eye and muscle symptoms.

    If you have stranges symptoms in your mother line, maybe it is worth checking it with a mito specialist?
     
  12. BeautifulDay

    BeautifulDay Senior Member

    Messages:
    310
    Likes:
    884
    Hello @aquariusgirl

    The MT-CYB rs3088309 which has been merged into rs527236209 is one of those spots with multiple issues where the scientists have screwed up the numbering on various instances and need to pull out the mis-labels and confusion before anyone takes it seriously. It's just part of being on the forefront of the infancy curve of genetics.

    For example, going into my 23andme results for MT-CYB, 23andme shows me as:
    "MT-CYB, rs3088309, position 15452, with the variants being A or C, and I'm an A"
    In fact, out of the 8 people I checked on 23andme, 7 of them have this mutation per 23andme. For example, it says both of my parents have it. Highly unusual. So that's an indication that it could be a mis-call or a position issue. Of course it could indicate my dad has that mutation also from the other side of the family. But the plot gets thicker.

    The problem is that if you go to the NIH site for the merged into rs# link, you'll find the position for this location is 15452. Yet, when I type rs3088309 into the search box on Enlis, it comes up with none of the 23andme data for that rs# per 23andme's raw data, but it does come up for one person with WGS for this rs#, except it's for position 15,453. So the position is moved over here by one.
    https://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=527236209

    Enlis has the Allele Freq for position 15,453 at 0.00% rare. Enlis has the Allele Freq for position 15,452 at 9.37% which is not so rare.

    Part of this screw-up is due to the old reference genome hg19 using a mitochondrial genome reference that was off in most locations by 1 or 2 (depending upon where it was) and also it has a few locations where the position was off. Most programs try to compensate for these issues. Yet, not all are picked up by all programs.

    While Enlis states that the allele frequency for this location is rare, if you go down to the bottom of the page for that NIH reference, it shows that it's not so rare. There is a lot of conflicting information because the original use of the wrong reference genome by everyone for years. Here is what the NIH has for position 15,453.
    https://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=527236184

    I had a Whole Genome Sequencing on myself (WGS) that called this position 15,453 as a mutation for me. Yet, they used the wrong reference genome hg19. When I backed it out of the original BAM file to the original FASTQ file and then reran it with the newer (revised reference genome) used by all the mito specialists, this mutations didn't show up.

    So to recap the position was merged. Bad sign. For rs3088309 (merged into rs527236209) it should show up as 15,452. Yet inputting rs3088309 brings up position 15,453 for a WGS on Enlis (but not those with the mutation per 23andme raw data entered into 23andme). This mutation was also an issue for the majority of companies that used hg19 as a reference genome for Mitochondria. Surprisingly, there are still some big companies still using reference genome hg19 today when doing WGS and WES, not realizing the issue with mitochondrial reference genome being off.

    Therefore, it's important to take this location with a grain of salt.

    Any location that has been merged will often have issues like this that will still need to shake out through the system and medical community. It's part of being at the forefront of genetics infancy.

    So please take this specific mutation with a grain of salt. A mito specialist would have to rerun it using a correct reference.
     
    pattismith likes this.
  13. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    Ok, thanks for the heads up.
    Can you say who did your whole genome testing? $?
    Actually, since you are here, can I pick your brain about a missense mutation that showed up?
    G1801394G MTRR missense...
    any thoughts on that one...
     
  14. BeautifulDay

    BeautifulDay Senior Member

    Messages:
    310
    Likes:
    884
    MT-ND5 rs28359178 is another spot where using the wrong reference genome has resulted in issues and should be taken with a grain of salt.

    23andme says I have "MT-ND5, rs28359178, position 13708, A or G, A"
    Running that through Enlis says I've got the the mutation and that 6.92% allele frequency.

    My WGS using the wrong hg19 reference genome again doesn't have that position, but has position 13,709 mutation which is listed as 0.00% allele frequency.

    This is another one of those spots where using that bad reference genome and not fixing it or not following through the whole pipeline on it can result in mis-calls.

    This is another spot I take with a grain of salt.
     
    pattismith likes this.
  15. BeautifulDay

    BeautifulDay Senior Member

    Messages:
    310
    Likes:
    884
    So you mean this one?
    https://www.snpedia.com/index.php/Rs1801394
    If it's that one, then it's very common and if I was you I wouldn't give it a second thought.
     
  16. BeautifulDay

    BeautifulDay Senior Member

    Messages:
    310
    Likes:
    884
    I like the company I initially used for a lot of reasons, but I won't recommend them until they fix that hg19 issue. I'm all in for supporting the adventurous companies out there, but since we are in the infancy, we too are adventurers. No company (even top hospitals aren't 100% error free in their pipelines). It's impossible to be perfect at this stage.
     
  17. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    Yes, that one. Yes, I see it's common. I need to check if my mom has it, but she is MTHFR compound heterozygous...with high homocysteine, so I still think it might be an issue for her.
     
    Learner1 likes this.
  18. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    yeah, this DNA thing is a bit of a dive down the rabbit hole at this point. I am not seeing a lot of return on time invested...
     
  19. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    I think that since our illness involves problems in sulfation and methylation... any weaknesses on those pathways just get amplified.
     
    Learner1 likes this.
  20. aquariusgirl

    aquariusgirl Senior Member

    Messages:
    1,584
    Likes:
    567
    Can you say what your expertise is in this field? Hobbyist? Undergrad, Masters, Clinical practice?
     

See more popular forum discussions.

Share This Page