Page 4 of 4 FirstFirst ... 234
Results 31 to 35 of 35

Thread: Playing around with qpGraph

  1. #31
    Registered Users
    Posts
    1,836
    Sex

    You guys shouldn't be using ancient samples from Harvard alongside those downloaded from elsewhere on the net if you want super accurate results.

    For instance, it's very unlikely that the WC1 sequence you have is really compatible with the Harvard samples, because it was genotyped differently. If you want to run WC1, than wait until it shows up at the Reich lab datasets page.

  2. The Following 2 Users Say Thank You to Generalissimo For This Useful Post:

     Awale (06-18-2017), Elizabeth (06-18-2017)

  3. #32
    Registered Users
    Posts
    3,809
    Sex

    Quote Originally Posted by Generalissimo View Post
    You guys shouldn't be using ancient samples from Harvard alongside those downloaded from elsewhere on the net if you want super accurate results.

    For instance, it's very unlikely that the WC1 sequence you have is really compatible with the Harvard samples, because it was genotyped differently. If you want to run WC1, than wait until it shows up at the Reich lab datasets page.
    In general good advise, but I have figured out a way to remap coordinates to the build used by Harvard (Hg19). So I just checked my WC1 bim file against the dbSNP database and verified that all my SNPs are mapped to Hg19. That is why when I merge files in plink I don't get the annoying error messages "multiple positions.....", and the merge goes smoothly. The reads are also mapped to the +ve strand, which is the more common convention.

    I also check with PCA after merger as an extra step.
    EurasianDNA.com - A study of the population history of West & South Asia.

  4. #33
    Registered Users
    Posts
    3,809
    Sex

    As an example, when I initially tried to merge the Reich Indian tribal Onge set with the Reich HO set in plink, I got something like 120,000 "same position" errors, and about 80,000 "tri-allellic" warnings.

    At this point some may be tempted to simply exclude the 120,000 SNPs and flip the 80,000 other SNPs. I don't think this is a good idea, because I checked the SNPs in the tribal set I was able to figure out that they were alligned with an old build from 2004, hg17.

    After I remapped the SNPs to Hg19 human reference, and retried the merger with the Reich HO set everything went smooth.
    EurasianDNA.com - A study of the population history of West & South Asia.

  5. #34
    Registered Users
    Posts
    291
    Sex
    Omitted

    Quote Originally Posted by Chad Rohlfsen View Post
    I'll add Vestonice16.
    Did that ever resulted in anything useful?

  6. #35
    Registered Users
    Posts
    1,937
    Sex
    Y-DNA
    R1b-DF27 FGC17112
    mtDNA
    H5a1

    Quote Originally Posted by epoch View Post
    Did that ever resulted in anything useful?
    Some interesting outputs, but I'm going to try something else first, and then I'll get back to it.

  7. The Following User Says Thank You to Chad Rohlfsen For This Useful Post:

     epoch (06-20-2017)

Page 4 of 4 FirstFirst ... 234

Similar Threads

  1. Replies: 6
    Last Post: 03-23-2015, 03:12 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •