No announcement yet.

Conflicting data in Chromosome Browser

  • Filter
  • Time
  • Show
Clear All
new posts

  • Conflicting data in Chromosome Browser

    I have attached an image from Chromosome Browser on Chromosome 11 with results from two siblings and a supposed cousin.

    The top bar shows the matching between Sibling 1 and the cousin, as well as with Sibling 2. The bottom shows the corresponding match between Sibling 2 and the cousin and Sibling 1.

    The segments showing the match between the two siblings are of course identical, and they show a long stretch in the middle where they do not match. Yet they both match the cousin for part of that stretch. I have labeled this area “Questionable Piece”.

    How is this possible? Here is my supposition: Sibling 1’s DNA is from the Paternal Grandfather (call it FF) and Maternal Grandmother (MM). Sibling 2’s DNA is from the Paternal Grandmother (FM) and Maternal Grandfather (MF). Thus Sibling 1 (FF/MM) does not match Sibling 2 (FM/MF).

    If the cousin’s DNA is either FF/MF or FM/MM, then the cousin matches both siblings, even though the siblings don’t match each other.

    As background, the cousin is a 2nd to 4th cousin match in Family Finder for both siblings. The cousin shares 58 cM with Sibling 1 (longest is 33 cM) and shares 49 cM with Sibling 2 (longest is 32 cM). The cousin does not have a tree available on ftDNA. The siblings are full siblings.

    Paper documentation shows a couple of geographic overlaps between the families of the siblings’ parents, but all impact 4G Grandparents or earlier. Of course there is always the possibility that there was another chance coupling downstream from there that could have connected the two families.

    Does my supposition make sense? Or is there another explanation?
    Attached Files

  • #2
    Have you lowered the default cM setting to show all segments 1cM or Greater or are you viewing at the default of only 5cM or greater?

    What is the end location of sibling share?
    What is the end location of cousin share?

    it maybe just that the extra cousin share is false depending on the number of SNPs tested between the two points

    you can find the number of SNPs tested between the two endpoints in your Build 36 Raw Data file


    • #3
      Here are the ranges using the default 5 cM threshold:

      Sibling 1 to Cousin: 19,077,087 to 59,330,556 (33.15 cM)
      Sibling 1 to Sibling 2: 188,510 to 46,717,489 (58.93 cM)

      Sibling 2 to Cousin: 19,077,807 to 58.358,885 (23.37 cM)
      Sibling 2 to Sibling 1: 188,510 to 46,717,489 (58.93 cM)

      If I lower the threshold to 1cM in Chromosome Browser, the ranges relative to cousin are unchanged, but I get the following for the Sibling ranges:

      188,510 to 46,717,489 (58.93 cM)
      48,014,889 to 56,273,717 (1.3 cM)


      When I compare the siblings in GEDMatch I get slightly different ranges:

      With threshold of 5:
      188,510 to 47,132,581 (68.8 cM) 13,465 SNPs

      With threshold of 1:
      188,510 47,132,581 (68.8cM) 13,465 SNPs

      I other words, GEDMatch doesn’t see the 1.3 cM range that Chromosome Browser does.


      At the very least, this approach does seem to eliminate the weird possibility of the cousin matching on both sides of my family tree.

      Great questions!