GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2006-01 > 1136433167


From: "Dale E. Reddick" <>
Subject: Re: [DNA] Whit Athey Y-Haplogroup Predictor Question
Date: Wed, 04 Jan 2006 22:54:55 -0500
References: <002001c611a9$06f5b8f0$0100a8c0@BigMem2>
In-Reply-To: <002001c611a9$06f5b8f0$0100a8c0@BigMem2>


Hi Folks,

John is correct.

I place far more trust in Whit's predictor than the predictor utilized
by FTDNA!

FTDNA predicted that I was haplogroup Q. However, eventually (after
nearly 10 months) - they found that I am R1b. This was more deeply
confirmed by Ethnoancestry when they found that I'm R1b1c.

Other than a SNP test, Whit's predictor is currently the standard by
which Y-STR results can predict how your haplotype fits into a
haplogroup assignment.

Dale E. Reddick
________________________________

John McEwan wrote:

>Dear Joe
>
>I was involved in a comparison with Whit's predictor, FTDNAs predictions
>as they appeared in Ysearch and a cluster analysis I did. All three
>methods use different techniques, but all are tied into SNP testing for
>comparison as well.
>
>Whits predictor is extremely good, I would not bet against it. My guess
>is for those groups that he predicts it is well above 99%. He recently
>found some unusual SNP results that conflicted with his predictor (and
>also my cluster analysis) and on extensive retesting these individuals
>were found to be wrongly SNP genotyped.
>
>Now this is not to say that Whits predictor (and method he uses) doesn't
>have quirks. In particular it has problems distinguishing haplogroup Q
>from R (well they are cousins) and this is identified by a reasonably
>high Q prediction in many R group individuals. However, that said I
>cannot say any more because you have not provided your Ysearch ID so we
>can examine the haplotype using other methods. It may be that it is
>pretty dubious and in fact is only 12 markers anyway. IMHO unless you
>have 37 STR markers you should spend your money upgrading to that number
>before considering spending money on SNP tests.
>
>Cheers
>
>John McEwan
>
>-----Original Message-----
>From: Joe Ragsdale [mailto:]
>Sent: Thursday, 5 January 2006 4:11 p.m.
>To:
>Subject: [DNA] Whit Athey Y-Haplogroup Predictor Question
>
>Question for the board. How accurate is the Whit Athey Y-Haplogroup
>Predictor? I ran my Values and this is what I got for the Values.
>Should I consider doing the Y-DNA SNP test from FTDNA?
>FTDNA suggests that I belong to R1a.
>As you can see the values are close on the Q nad R1a Haplogroups.
>
>
>Regards
>
>WCR
>
>
>
>
>
>
>
>How Well Does the Haplotype Fit the Following Haplogroups (Scale of 0 to
>100):
>
>
>E3a E3b G I1a I1b I1c J2 N Q R1a R1b
>5 12 2 2 10 4 6 1 28 34 13
>
>Values above 50 indicate a good fit with the haplogroup. Values between
>20 and 50 indicate a fair fit.
>Values below 20 indicate an unlikely membership in that haplogroup.
>
>

This thread: