y DNA

[stonechat]

I have still not taken the plunge
I will probably go with FTDNA
I will get autosomal and y DNA tests
But how many markers should I go for on the Y DNA test?

[rlw254]

Many people 'go big or go home' and go for at least y67 or more to start. I started with y12 and then opted to look at SNP testing from there. I would advise joining a surname project if one exists for your name and talk with the administrator to see what their group members have traditionally done to yield the most information.

[clayton bradley]

For Y-DNA I started with 37, joined the project for the haplogroup and then bought individual SNPs to progress. I hope you do autosomal, Stonechat, since you may be distantly related to my husband. It could be interesting to see if you come up as a match on FTDNA or Gedmatch, if you upload to Gedmatch, cb

[RobertCasey]

I really believe that 12 markers (80 to 90 % error rates) and 25 markers (40 to 60 % error rates) tests should really be avoided as the false positive rates are just to high. They are useful if you have researched a line that you believe could be related and just want to see if there is a chance of being related. 37 markers (20 to 30 % error rates) is the first resolution to have reasonable false positive rates. But 67 markers is definitely the sweet spot: 1) it reduces the false positives by another 10 % or so (improvement but not radical); 2) It allows for YSNP prediction of branches in the 1,500 to 2,500 year range with over 95 % accuracy which is really great for haplogroup research (you may be able to eliminate a tier of SNP pack testing as well); 3) the least know advantage for many haplogroups, 67 markers and extensive YSNP testing allows charting of connections which really helps with advice for next steps for YSNP testing. You really have to belong to haplogroup that is very prolific (or very well YSTR tested as well) AND have a ton of Big Y tests yielding dozens of branches under a predictable YSNP in the 1,500 to 2,500 year range.

http://www.rcasey.net/DNA/R_L226/Haplotrees/L226_Home.pdf

I am not a big believer these days for 111 marker upgrades - unless you find out very soon that you have many closer matches - then it helps. In the next year or two when the read lengths of NGS testing reaches 500 to 1,000 base pairs, 111 marker tests will no longer be required as you will be able to reliably extract out all of the FTDNA 111 markers in the new future from NGS tests. Once that happens, 500 YSTRs will be used vs. 111 markers (but this resolution will require some very costly software to properly analyze with so many markers involved).