GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-11 > 1132321556


From: "Ken Nordtvedt" <>
Subject: Re: [DNA] Predicting haplogroups from haplotypes
Date: Fri, 18 Nov 2005 06:45:56 -0700
References: <20051118063559.93528.qmail@web50710.mail.yahoo.com>


David, Your tirade is late; I was expecting it sooner after my post.
Unfortunately my mouse ball is worn out, so a detailed reply must wait for
the computer store to open up so I can get a new part. With your volume of
email I bet this has happened to you. Anyway; great news about the new
haplogroup I SNPs, and I will have something to say about them as well.
Congratulations to EA for seeking new SNPs and to Gareth and Jim for their
lab prowness in finding them.

Ken
----- Original Message -----
From: "David Faux" <>
To: <>
Sent: Thursday, November 17, 2005 11:35 PM
Subject: [DNA] Predicting haplogroups from haplotypes - Fraught with
Pitfalls


> Are you kidding here Ken? Are you truly serious? Is this your advice to
> newbies?
>
> Can you predict the A1 our multiplex identified or the difference between
> the C* and C3 reported on the same day - based on haplotypes with extended
> markers? How are you on Q* versus Q3 on the basis of a haplotype? Can
> you tell a J1 from a J2 haplogroup from any number of Y-STR markers?
> Considering R1b, can you tell what a SRY2627 haplotype looks like (this
> haplogroup cannot be predicted on the basis of Y-STR markers of any
> number). How about S21? Is it just your Frisian type that is S21+? If
> you think so you would be entirely wrong - we have haplotypes from Orkney
> that aren't even remotely close. You are going to run into many
> haplotypes you have wrongly predicted but just don't know it yet. I have
> a bunch of M284 positive that are hanging in mid air. I don't know what
> to tell them other than my advice is to start with a marker such as M170
> and move downstream and see what turns up. Do you think that they are all
> M223 derived? How would you know unless you enco!
> urage
> them to test - you are otherwise making assumptions and yet on the other
> hand say you want to find odd clades but still will not recommend that
> people take the SNP tests that would identify these. Do you really expect
> that a company is going to want to work with someone and develope primers
> and put in thousands of dollars and hundreds of hours of work testing a
> SNP only to tell people that they don't need to bother with it anymore
> because you can predict it from a Y-STR score if people would only buy
> more markers. If fact that is all very ill - advised as in fact the only
> true way to determine a haplogorup is with a SNP test. A prediction
> provides the foundation or basis upon which to chose a series of selected
> SNP tests such as those that fall under I1c. We just found two new SNPs
> in the I category - can you predict what haplotypes these will have? Do
> you recall the P19+, M170- that we found - did you predict that. I think
> not, it caught everyone by surprise. Withou!
> t a SNP
> test no matter what someone predicts it is not acceptable to insert this
> in a SNP database unless it has been tested. There are always surprises.
> Convergence happens like it or not.
>
> David Faux.
>
> Ken Nordtvedt <> wrote:
> But what bothers me is that many entrants into the hobby don't know
> (aren't
> told) that if they would purchase extended STR haplotypes, they probably
> don't need any SNPs. The extended haplotype in the bulk of the cases and
> with high probability will indicate haplogroup affiliation. Only a small
> number of haplotypes are ambiguous if extended (many markers).
>
> Ken
>
>
>
> ----- Original Message -----
> From: "Robert Stafford"
> To:
> Sent: Thursday, November 17, 2005 4:15 PM
> Subject: Re: [DNA] re: SMGF Search Technique
>
>
>>I would go with the sure thing, rather than taking a chance. Most will
>>want
>>20+ markers anyway. Now people can get a SNP plus 26 markers for about the
>>same price as FTDNA's 25-marker test alone. Simply use the SMGF coupon for
>>a $95 26-marker test and order the Marligen test at EA for $85 (or $99 at
>>DNAH). The multiplex gives a lot better resolution and appears to be more
>>accurate than FTDNA's SNP test.
>>
>> If a person also wants DYS464, he can order an upgrade at RG for $75 or
>> order a DYS464 test along with other unique markers at DNA-Fingerprint.
>>
>> Bob Stafford
>
>
> ==============================
> Jumpstart your genealogy with OneWorldTree. Search not only for
> ancestors, but entire generations. Learn more:
> http://www.ancestry.com/s13972/rd.ashx
>
>


This thread: