Archives Search Engine

Body:
Subject:
From:
(email address of poster)
List:
(limit search to one mailing list)
Date:
(e.g. 10 Jun 2005, Jun 2005, or 2005)
   Search tips

Searching for: +path:genealogy-dna +(+date:aug +date:2002)
Viewing 1-25 of 613 matches from 36,014,699 documents1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | Next
1. Re: [DNA] Help with REO results FTDNA [1]
Jim wrote: > If you analyze this group's data by the following: > Obtain the absolute count difference my count each locus for each person, > the result is an array of five columns 19 rows of ones, some twos and threes > a smattering of fours. Assuming the hypothesis that zero difference > translates to a common ancestor, then one way analysis of variance on > columns is whether a column or any column is significantly different from > zero.... > analysis of variance required no assumptions of > mutation r
2. Re: [DNA] Help with REO results FTDNA [1]
I wrote: " you simply take the number of steps and divide by the number of loci and by the mutation rate to get the expected number of generations." Jim replied: > Please show me an example calculation: > (1) In a DNA test of 23 loci I was identified as having a common ancestor > with another whose DNA and mine differed as: > DYS 390 mine was 23, his was 24 and at DYS 392 mine was 11 and his was 13. Are you asking how many steps this would be? As a rule of thumb I generally treat an isolated 2-step jump
3. Re: [DNA] the value of 12 markers [1]
Mitzi wrote: > I have been proselytizing a couple of Clarks in my line to be a part of the > Clark surname test. It's being done with 12 markers unless one specifically > asks for 21. That's a little out of date. FTDNA has retroactively upgraded all their old 21-marker results to 25, and all the newer high-resolution results have 25. > Is this going to be worth it? Will we learn that much more if we > kick in $59 more for the 21 marker test? The price differential has also increased -- to $70. > Our
4. [DNA] Mutation rate [1]
All this discussion of the STR mutation rate reminded me of the various proposals made in the past to determine that rate from the ever-growing collection of DNA studies among us. Basically, when Allan Gleason tried to gather statistics by looking at various web sites, he discovered that he couldn't because the necessary statistics aren't shown and couldn't even be deduced. The only exception, in fact, was Allan's own site, which included a labeled pedigree chart. I made a similar survey of my own a whil
5. Re: [DNA] RICE OR ROYCE DNA [1]
Ruth wrote: > A few days ago someone by the name of RICE posted a message that he had been > informed by a lab testing his DNA that he was really a ROYCE. I suspect that's a paraphrase that lost something of the original. For one thing, those two spellings were virtually interchangeable in the 16th century (when surnames were young), and there was considerable jumping back and forth even later. If you want to see the background, go to the Edmund Rice Association web site http://edmund-rice.org an
6. Re: [DNA] How specific is 12 marker test [1]
Sanford wrote: > I have the results of my 12 marker tests and noted that there are at least four > people that match my profile 12 of 12. However these people all have different > surnames from me and from each other and as far as I can tell are not related. > If mutations have occurred at such a rate that people who are not really > related match my profile, then how valuable is the 12 marker test? It's not necessarily that. Mutations happen at random, so you could just be one of those "lucky" fami
7. Re: [DNA] Re: GENEALOGY-DNA-D Digest V02 #355 [1]
Iris wrote: > I see it noted that a reading of 13 on DYS 392 is one of the defining > readings of Haplogroup 1 No, it doesn't DEFINE hg1. Rather, it is often observed in people who belong to hg1. > that Haplogroups 2 and 3 often have 11. > What about a reading of 14 on DYS 392? Anyone have comments or > information on that? Indeed, 14 is only one step away from 13, so 14 also must be fairly common in hg1. Perhaps only 1/10 as common, but that's still a lot. Back when the European YSTR database had
8. Re: [DNA] Re: GENEALOGY-DNA-D Digest V02 #296 [1]
Mary wrote: > If an 85-year-old man has had 12 blood transfusions while in his 70's, and > several platelet transfusions while in his 80's, can this account for his > 25-marker test showing TWO mutations from a third cousin? No. The tests are normally done with mouth cells, not blood. Even if a blood sample were used for the DNA test, it would show no trace of transfused cells from many years earlier. Indeed, if a blood sample were tested soon after a transfusion, the result would likely just be the dre
9. Re: [DNA] CMH haplotype [1]
Iris wrote: > I've been reading the posts for several weeks now, but still haven't > seen a discussion of the following. Can anyone assist me in > understanding what 6 chromosomal markers make up the Cohen Modal > Haplotype (CMH). CMH is DYS393=12, DYS390=23, DYS19=14, DYS391=10, DYS388=16, DYS392=11. > I've seen mention of the YAP--which marker is that > and how does one know if it is indicated in one's results? That's a UEP (Unique Event Polymorphism). It is not one of the markers tested by the comm
10. Re: [DNA] Request for Haplotype/Haplogroup analysis [1]
Steve wrote: > Using the 6 marker Y DNA set of DYS388, 393, 392, 19/394, 390, 391 our most > recent results are 10, 13, 11, 16, 25, 10 and 12, 13, 11, 16, 24, 10. Can > someone type those to haplogroups? The second one is only two steps away from the typical pattern of haplogroup 3. It might be that. The first one is three steps away and is thus more iffy. John Chandler
11. Re: [DNA] Y haplogroup distributions [1]
Nick wrote: > John Chandler wrote: > Viking does NOT imply hg2, and hg2 does NOT imply Viking. (Here, I am > using the word "Viking" as shorthand for "Scandinavian origin".) > In a western European context, hg3 apparently does imply Viking, but > Viking does NOT imply hg3. > Viking is definitely a mixed bag. > > John I respect your grasp of the technical details and we have argued > about this in the past but you are making too strong a statement here > about HG2. Not at all. I made precisely the right s
12. Re: [DNA] Celts [1]
John wrote: > I am curious to know whether any tests have derived a haplogroup for the > Celtic people. Unfortunately, as in the case of almost any oversimplified question, the answer is "yes and no". You are assuming that there is, in fact, such a thing as "the Celtic people" with some sort of common genetic heritage. You have to realize that the the term "Celtic" is linguistic, not ethnic. It may be convenient to assume that speakers of Celtic languages have something genetic in common, but you have onl
13. Re: [DNA] mutations at 385a and b [1]
Don wrote: > Can someone help me to assess a "single" mutuation between 2 living Maxwell > males, with ancestors from same locale, era, at least 8 generations > seperated, but not "proven" to be related . > Specifically on 12 loci Y chromosome there is a 11/12 match except: What you really need is a combination of DNA evidence with whatever else you can muster. For example, you could study all the other Maxwells from the locale and era and try to figure out where they came from and whether they were relat
14. Re: [DNA] Wingfield Y-DNA [1]
Lee wrote: > X Wingfield is paper-documented from his Great-Great Grandfather in Group-A > (I have all his believable documentation in my possession). > > However, X matches only 2 of 24 (that's II of XXIV) with our main testee from > Group-A. Are you saying that you have solid paper documentation for both, linking them to a shared gggf? If so, that's recent enough that you can probably rule out an unsuspected adoption, so you would likely conclude that AT LEAST ONE of the two has another kind of previous
15. Re: [DNA] Re: GENEALOGY-DNA-D Digest V02 #290 [1]
Hello, everybody. I'm back. I see there has been a lot of discussion in the meantime, so it will take me some time to get caught up. Mary wrote: > Dear John C., > > Are you saying that if I got hair ROOTS when I collected my sample, I could > successfully sample the Y DNA? But if I get hair WITHOUT roots, I could only > have a mt DNA analysis? In principle, yes to both. In practice, it might depend on how many hair roots were included. Success in extracting mtDNA might also depend on what treatment t
16. Re: [DNA] Locus values [1]
Gary wrote: > From BYU (Relative Genetics) I received 23 marker test results. That's fascinating news! In the recent past, the number of markers has been consistently 24. Did your report of 23 come just now, or was it many months ago? A month or so ago, there was a paper that showed the 24 markers were actually only 23 plus a duplicate, so it would be interesting to know if the duplicate has now been dropped from the list. What loci are included? > Four of the alleles have values outside of the nomenc
17. Re: [DNA] Help with REO results FTDNA [1]
John wrote: > I note the MCRA calculator does 0,1,2 steps > I would like to put two items on the WIBNI [ Wouldn't It Be Nice If ] > list for the next version of the MRCA > > An input for the number of steps different. > Reason 1: I am getting and seeing in the List were 3,4,5 steps are at issue > Reason 2: In my first two surname results we had 7 steps out of 23 When you look more closely, the cases where 3 or more steps are "at issue" are not really serious cases -- more a matter of wishful thinkin
18. Re: [DNA] Wingfield Y-DNA & Watkins Y-DNA [1]
Lee wrote: > Wingfield is my wife's family. I have no > Wingfield blood. But, taking a look at Table 2 below shows that my Y-DNA > matches 19 of 24 loci of two testees of the William & Mary (Messer) Wingfield > bloodline. That is a "closer match" than any of the three > Wingfield/Winfield lines so far tested match each other! With two fewer > non-matched loci, I could have been a Wingfield -- !?@?#%*?+?....? Not really. With 3 discrepancies out of 24 loci, you would still be marginal at best, ev
19. Re: [DNA] Help with REO results FTDNA [1]
Jim wrote: > No sir... I'll make one last effort to show you what's wrong with your method. Here's a very simple example to illustrate. Suppose you weigh a man every day, not necessarily the same man, and you want to know which measurements are the same man and which aren't. Unbeknownst to you, it is indeed the same man every day, but, also unbeknownst to you, the scales start to malfunction after the third day and give you a weight that is off by an amount that depends on the barometric pressure. The s
20. Re: [DNA] Help with REO results FTDNA [1]
Jim wrote: > Again No sir: My error estimates are not arbitrary. They are based upon > three and higher order interactions, which are ordinarily assumed equal to > zero, any estimate different than zero are error. This is usual assumption > with data analysts and places the burden of argument elsewhere. The argument > can make progress via additional data sets, as I have said more than once. The problem is that there are many assumptions here, and several of them are unwarranted and thus arbitrary. For st
21. Re: [DNA] Help with REO results FTDNA [1]
Jim wrote: > Relative Genetics and they are > confident in their conclusions, but they use .004 as the mutation rate. There you have it. They assume a mutation that's twice as big as anyone else uses. That's the same as dividing all the mutation distances by two. Certainly, if you cut down the distance from 3 steps to 1.5, it would make a big difference. However, they haven't presented any evidence to justify their assumption. John Chandler
22. Re: [DNA] MRCA calculations in reverse [1]
Earl wrote: > Your calculations are based on the assumption that changes of a marker > occur on average at a rate of about 1 in 500 transmission events. > With 25 markers the rate is about 1 mutation per 20 transmission > events. Implied is that brothers have haplotypes different by one > mutation about 10% of the time. A three mutation difference between > brothers would certainly be cause for concern. But with second > cousins this simple estimate says that three mutation differences > happen about 6%
23. Re: [DNA] Help with REO results FTDNA [1]
Jim wrote: > No Sir, the line is not arbitrary. Whether one column is zero or something > else depends on error estimated from 3 and 4 factor interactions. The problem is that your method of estimating errors is indeed arbitrary. You have necessarily used the assumption that the variation displayed in your sample is an intrinsic measure of "error". Unfortunately, your sample is not only too small to be useful for this kind of exercise, but also of necessity biased. Ultimately, the statistical tools you a
24. Re: [DNA] MRCA calculations in reverse [1]
Ann wrote: > I interpreted Earl's question differently than John. More than that, she remembered to multiply by two for the MRCA, which is something I forgot to do in this little exercise. I estimated the 95% confidence lower limit for 3 mutations as "26", but that was transmission events, and I treated it as if it were generations. Sorry, all. In other words, 20 generations *would* be within the 95% confidence interval for 3 mutations. Nonetheless, I'm going to stick with my opinion that 3 mutations i
25. Re: [DNA] Help with REO results FTDNA [1]
Earl wrote: > Let's assume that the two men have an MRCA at, say, 20 generation. > The probability is not high that they would test 3 mutations apart. > But having found it, is the probability so low that you would reject > the assumption? Especially if other evidence is relatively > convincing? A) Yes, the probability is very low (outside the 95% confidence interval) B) 20 generations is likely to be in the early 1400's -- I rather doubt you will find "relatively convincing" evidence anyhow.
Viewing 1-25 of 613 matches from 36,014,699 documents1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | Next

CPU seconds used 0.569913