www.dna-forums.com dzial gdzie omawiana jest FTDNA.
The raw data comes in two separate files, one for autosomal (chromosomes 1-22) and a separate file for the X chromosome. Apparently no mtDNA or Y chromosome.
Both files are in TSV format which can be opened with Excel or Notepad, etc. and it has one header line:
RSID CHROMOSOME POSITION RESULT
rs123 1 12345 CC
rs456 2 56789 --
Positions are based on build 36 reference. I'm assuming that the results uses plus orientation (forward strand) but haven't verified.
(I'm not sure the header for the fourth column as I asked him to delete that column so he wouldn't have to reveal his results. I've asked him to check what it is and will update.)
Now the interesting part for those who have tested with 23andMe or deCODE and wondering if it is worth testing with Family Finder or will it just be a duplication?
Total SNPs:
23andMe v2 (2010-03-26 revision): 578,320
deCODE v1 (as of 2008-09-30): 1,013,349
Family Finder: 563,800
23andMe and Family Finder combined: 994,075
Comparing 23andMe to Family Finder:
SNPs in common: 148,045
SNPs unique to 23andMe: 430,275
SNPs unique to Family Finder: 415,755
Verdict: Surprisingly little overlap between the two.
Comparing deCODE to Family Finder:
SNPs in common: 209,099
SNPs unique to deCODE: 804,250
SNPs unique to Family Finder: 354,701
Verdict: A bit more overlap than with 23andMe, but still a lot more genome to explore.
And to round out the comparison:
Comparing 23andMe to deCODE:
SNPs in common: 554,618
SNPs unique to 23andMe: 23,702
SNPs unique to deCODE: 458,731
Verdict: If you've already done deCODE, testing with 23andMe doesn't add much. If you've tested with 23andMe, adding on Family Finder or deCODE both add similar numbers of SNPs, and Family Finder is a lot cheaper.
Now let's say you've testing with 23andMe and Family Finder and wondering if there's any reason to add deCODE into the mix:
Comparing 23andMe plus Family Finder to deCODE:
SNPs in common: 618,051
SNPs unique to deCODE: 395,298
SNPs unique to 23andMe plus Family Finder: 376,024
Verdict: You can still add another 395k SNPs, but is that worth $2000?
The above comparison don't consider whether the SNPs tested are actually interesting or not, just the number of them tested, but they do show that there is relatively little overlap between the three tests.
The SNP lists and comparison lists are all available here: http://vps1.jameslic...na/snp-compare/
misty69