1. Patients launch $1.27 million crowdfunding campaign for ME/CFS gut microbiome study.
    Check out the website, Facebook and Twitter. Join in donate and spread the word!
Hyperparathyroidism: An Often Overlooked Differential Diagnosis to ME/CFS
Andrew Gladman puts hyperparathyroidism under the microscope, exploring what the disease is, how it can mimic ME/CFS in presentation and how it is treated.
Discuss the article on the Forums.

Download for Rare SNP Analysis

Discussion in 'Genetic Testing and SNPs' started by Valentijn, Aug 29, 2013.

  1. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
    Valentijn Everything working fine. At first I thought a problem with the out put but then realized that the reason I only have 3 pages of genes and my aunt has 6 is she has more 1%ers. Everything worked as described once Java installed. Easy Peasy. Good Job. Now I need to go look at what they are. I had 3 homozyg and my aunt had 6. Thanks to you and your genius programmer associate ;)
    Valentijn likes this.
  2. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
    Do we think the hetero's are of importance? Going to that other thread you linked above.
  3. Valentijn

    Valentijn Activity Level: 3

    Messages:
    6,041
    Likes:
    8,401
    Amersfoort, Netherlands
    It really depends on the circumstances ... in some cases, problems can be dominant instead of recessive, in which case one copy is enough. And two different missense mutations on different strands (one from the father and one from the mother) of the same gene can certainly cause big problems, just as much as a homozygous mutation sometimes.

    If a problem is dominant and hetero, it should be apparent in the parent with the same allele, going back for however many generations. So those tend not to cause the serious "stealth" problems like you'd expect for a genetic involvement in something like ME/CFS.

    But it's also possible to have, for example, one rare "recessive" heterozygous mutation on one SNP which is interacting with a less rare mutation of another SNP on the gene. So basically, we can't really be certain, which is why it's nice to look at all of the rare results, for purposes of comparison with other ME/CFS to see if there's any shared rarities.
    roxie60 likes this.
  4. Valentijn

    Valentijn Activity Level: 3

    Messages:
    6,041
    Likes:
    8,401
    Amersfoort, Netherlands
    We've added another database, which generates a list of all SNPs you have which are at 10% or less prevalence rates. This file's 10 times bigger than the 1% file, at 3.5 MB. It generated about 18,000 results for me, so it's far less useful than the 1% file if you want to look rare genes up and learn more about them, etc.

    But what it is good for is generating a big list of rare homozygous results. Basically the 1% file is only going to show homozygous results present in 0.01% of the population or less. But by looking at genes in the 10% or less range, you can generate a big list of homozygous results (which are much more likely to cause big problems than heterozygous results) which are prevalent as homozygous results in approximately 1% of the population or less.

    So you can generate the results, save as text (PDF would be HUGE), open it in Excel, sort according to the ETC column (which flags homozygous results), and delete everything after the homozygous results stop. This is still a big list - for me it produced 950 homozygous results. But if organizing by %, you can still use it to look up the rarest results, such as I've done for my 75 rarest homozygous results with prevalence ranging from 0.01% to 0.16%.

    Another application could be to contribute our results to a shared list to determine shared rare homozygous variations among ME/CFS patients.

    Anyhow, the 10% file is at http://sourceforge.net/projects/analyzemygenes/files/Databases/ . If you already have the program, you just need to click on ten_percent.zip to get the new file. If you don't have the analysis program yet, you should first download that separately at http://sourceforge.net/projects/analyzemygenes/ by clicking the green button.

    To use the 10% file, you'll have to select it manually when running the genes program. You can do that by clicking on the top button (which automatically selects the 1% file), and selecting the ten_percent file instead.
    merylg and roxie60 like this.
  5. Critterina

    Critterina Senior Member

    Messages:
    672
    Likes:
    276
    Arizona, USA
    Hi Val,
    I downloaded the new file, unzipped, and ran it. When I hit analyze, the screen popped up, I could see the Save As buttons, they were not grayed out anymore, but the status bar never showed up. I waited 2 minutes then tried to save as PDF. I gave it a name and location, but it did not create a file.

    I saw that you had new 1% and 10% files, so I downloaded them. I put them in different folders. Then I tried swapping out the 1% file (one_percent.h2.db) and rerunning the program. Same thing happened.
    Valentijn likes this.
  6. Valentijn

    Valentijn Activity Level: 3

    Messages:
    6,041
    Likes:
    8,401
    Amersfoort, Netherlands
    Did the one_percent file get unzipped? I can recreate the same situation: it lets me "analyze" the zipped file but doesn't do anything and produces no results.
  7. bel canto

    bel canto Senior Member

    Messages:
    188
    Likes:
    170
    Hello, Val - the 10% download worked apparently perfectly, and I was able to save the results in different formats. The only thing I noted is that you were describing something that would show up in red, and my results are black and white only. Am I missing something?
    Valentijn likes this.
  8. Valentijn

    Valentijn Activity Level: 3

    Messages:
    6,041
    Likes:
    8,401
    Amersfoort, Netherlands
    It will only show up in red if stopping the program (by closing the progress bar) before it's done, to let people know they didn't generate full results.
  9. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
    10 percent done, 979 homozyg, none less than 1%
    Valentijn likes this.
  10. Valentijn

    Valentijn Activity Level: 3

    Messages:
    6,041
    Likes:
    8,401
    Amersfoort, Netherlands
    If you want to send me a list of just the basic RS (and "i") numbers from the first column, I can run them through excel to compare to mine. A PM in my Phoenix Rising mail might be the best way we can avoid spamming everyone with the full lists :p
  11. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
  12. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
    979? or just a subset?
  13. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
    Valentijn ck your email
    Valentijn likes this.
  14. roxie60

    roxie60 Senior Member

    Messages:
    1,636
    Likes:
    493
    Central Illinois, USA
    researching genes/SNPs is always a lot of work but this 10 percent file that produces a list of at risk/rare SNPs helps to provide focus. Interesting some of my homozyg research already related to folic acid, FIGLU, Glutamine.
    Valentijn likes this.
  15. ukxmrv

    ukxmrv Senior Member

    Messages:
    3,330
    Likes:
    1,762
    London
    Thanks Val!,
    all looking good for me. I'm just comparing the new one % results file generated this way with the earlier one I generated using that earlier online one.
    Valentijn likes this.
  16. bel canto

    bel canto Senior Member

    Messages:
    188
    Likes:
    170
    Val - if you want my 1% and 10% rs list, please let me know how to send it to you.
    Valentijn likes this.
  17. Sparrowhawk

    Sparrowhawk Senior Member

    Messages:
    447
    Likes:
    228
    West Coast USA
    Bloody great idea, across the board. I wonder what it would take to have a database of gene info for folks on the forum, and then sift for specific variations. All crowdsourced and informal, mind you, but what if we found something in common no one has seen before? Am I just ignorant of the amount of work this might take?
    Little Bluestem likes this.
  18. Sea

    Sea Senior Member

    Messages:
    683
    Likes:
    719
    NSW Australia
    I can't get it to work for me. I'm on a Mac running 10.6.8 and have Java 6. Any ideas why that won't work?
  19. bel canto

    bel canto Senior Member

    Messages:
    188
    Likes:
    170
    Sea - I'm on 10.6.8 and I have it working
    I'm not an expert, but if you want to pm me about what's happening when you try it, I'll try to help.
    I think that my Java is 13.9.7. Are you current on your software upgrades?
  20. Sea

    Sea Senior Member

    Messages:
    683
    Likes:
    719
    NSW Australia
    Thanks for your input, it's good to know it's possible on the Mac 10.6.8.

    Yes I'm current on software upgrades. I thought Java 7 onwards was only for Mac 10.7 and later. I'll look further into the different versions of Java.

    I can download the gene program and see the three files but it won't open.

See more popular forum discussions.

Share This Page