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:feb +date:2003)
Viewing 1-25 of 686 matches from 36,222,914 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] Y - STR European Database: Minimal Versus Extended Haplotypes [1]
David wrote: > The moral here is that those of us with a common haplotype (e.g., the Atlantic > Modal Haplotype) are going to have to wait a very long time before we can > derive any useful information from the Y - STR database. I disagree. It really comes down to what you are willing to call "useful". You have learned approximately how commonly your haplotype occurs throughout Europe. You may call that useless if you like, but I don't. The database was never intended to provide anything more than th
2. Re: [DNA] News item: SOPER DNA project [Rice] [1]
Ann wrote: > seriously entertain the possibility that somebody goofed during one of the > transcriptions and stuck Abishai and Matthias under the wrong set of parents. As it happens, Abishai also appears in the New London VRs, all by himself, listed as son of Gershom (no mother mentioned). Bear in mind that Groton wasn't set off as a separate town until 1705. Also, when Gershom moved to Worcester, the children mostly followed, and Gershom deeded land there to son Matthias. There is also a letter from Ge
3. Re: [DNA] mutations and paper trail [1]
Nel wrote: > First, I'm not quite sure why John is interested in the lineage/transmission > events of other surname family tests.... > Unless I'm reading something wrong, you cannot calculate > transmission events between 2 individuals unless there is a known common > ancestor. Let me clarify. First, the transmission events are the basic ingredient in mutations. You can't calculate a mutation rate without knowing two things: the number of mutations and the number of mutation opportunities. (If everybody
4. Re: [DNA] mutations and paper trail [1]
Pieter wrote: > But, we disagree. I highly respect your opinion, but my remark is not really a > statiscal statement: I merely look for other experiences in one name surname > studies with mutationrates higher than the rule 1 /500 generations per > marker. Ah, but you can't have it both ways. If it's not intended as a statistical statement, you must not attempt to draw statistical conclusions from it. What you could truthfully say was that you have observed a higher than "expected" average rate in you
5. Re: [DNA] Testing [1]
Bob wrote: > I would stongly recommend that anyone using the tests for genealogical > purposes (typing a tree) go with the 25-marker initially. You will find 12 > markers are not fine enough, since you are hoping for some participants with > a single mutation that marks a branch. I disagree. Mutations are not desirable. Think about it: the only way to make a catalog of all the mutations that "characterize" the various branches of your tree is to test ABSOLUTELY ALL male-line descendants of the founder.
6. Re: [DNA] DNAPrint - AncestrybyDNA BGA Testing ("Ybase Type") Database [1]
Ann wrote: > I do have some reservations about the methods he used in the section he > called "Ancestry Limit for Gene Transfer to Descendants." I need to think > this through, but I'm not sure that it's appropriate to think about 60,000 > genes (quite aside from arguments about how many genes there are). You > inherit (or lose) the DNA of your parents in big chunks, which are smaller > than chromosomes because of the cross-over effect, but larger than genes. > John C, do you have any thoughts on this? I
7. Re: [DNA] Marker equivalent [1]
Julian wrote: > I need to know the equivalent of markers between: > > Anknown laboratory 25 numbers > 393 390 19* 391 385a 385b 426 388 393 389-1 392 389-2 458 459-a 459-b 455 454 > 447 437 448 449 464-a 464-b 464-c 464-d > > Relaive genetics. 23 numbers. > 385a 385b 388 389 389II 390 391 392 393 394 426 437 438 439 462 GGAAT1BO7 YCAIIA > YCAIIB Y-GATA-A10 460 461 Y-GATA-C4 Y-GATA-H4 Well, some of these have exactly the same labels, as you can see: 385a=385a, 385b=385b, 388=388, and so on. These should
8. [DNA] Kit numbers [1]
I have just made an addition to the Blanchard DNA project web page that might amuse you all. For more than a year now, we have grown accustomed to seeing results from FTDNA presented with handy little 4-digit indentifying numbers. I have made a list of all the kits used in the Blanchard study along with their dates of issue, added in a few more kits from the Blakesley, Duerinck, Kincaid, and Rice studies to fill out the time coverage, and made a graph showing kit number vs time. (Thanks to the other stud
9. Re: [DNA] News item: SOPER DNA project -- everyone please read [1]
Ann wrote: > If some of you have had 25/25 matches with different surnames, could you > share your decisions about whether or not to pursue the genealogy? What > factors did you consider? The Rice study encountered an extraordinary coincidence -- so amazing that we are still trying to convince ourselves it really *is* just a coincidence. When we started encouraging participation by people from lines other than our main focus (descendants of Edmund Rice), we naturally included people with the surname ROYC
10. Re: [DNA] mutations and paper trail [1]
Orin wrote: > For what it is worth, Relative Genetics is using a higher number of 0.003 Note that RelGen was using 0.004 last year and hotly defending it. It was said to be based on "in-house" (i.e., never published) data. I have to wonder what the standard deviation was for the old estimate and what it is now for the new one... John Chandler
11. Re: [DNA] Y - STR European Database: Minimal Versus Extended Haplotypes [1]
Alan wrote: > Using 11 markers and the Poisson constant of 0.2% per expected mutation per > marker = 11x 0.2=2.2%, a one step change = 100%, 100% divided by 2.2% = > 45.45, this being the expected number of transmissions when a mutation would > probably occur. Taking 25 years as a generation , 45.45 x 25 = 1136 years. Ann has meanwhile answered the questions she posed as possible ways to look at the problem, but I think it may be instructive to point out the good and the bad in Alan's calculation as quoted
12. Re: [DNA] mutations and paper trail [1]
Pieter wrote: > Based on latest results in my project( see previous message) I do think that > Mutatons in Y DNA testing are more frequent that the famous rule of thumb 1 per > 500 generations (per marker) makes us believe. The above is, or perports to be, a statistical statement. However, the reasoning behind it turns out to be what we in the business call BIASED statistics. Any time you focus on one particular case, you magnify the effect of the normal fluctuations. When the average rate of a rando
13. Re: [DNA] FTDNA Haplogroup Database [1]
Dan wrote: > My question: Cannot the same group analysis > which was used to create the archaeologically interesting haplogroups be > also used to derive a phylogenetic tree based on the STR counts? This might > approximate a "family tree of surnames", although genetically based rather > that sociologically based. Are there just too many variations and not > enough samples? The problem is more fundamental than that. Yes, there are a lot of variations, but it goes beyond mere numbers. SNPs are so rar
14. Re: FW: [DNA] mutations and paper trail [1]
Bill wrote: > In our LOWES, we have documented each descendant and found > on men with a 1 marker mutation (389-2) out of 12 in the last 142 years > .not 500 years as predicted...... At this rate we would have about > 3.52 mutations in 500 years in that time.... You're missing a very important point. There is no single steady rate of mutation in your study or anywhere else. There is an ACTUAL rate that happens to be exactly zero most of the time, but it jumped to one for that particular marker at a parti
15. Re: FW: [DNA] mutations and paper trail [1]
Bill wrote: > Lets fill YBASE up with test results and then each of us can begin looking > at > the various rates mutational changes between the markers of related clans or > SURNAMES or haplotypes.... Note that the level of supplementary information suggested in YBASE (i.e., the name and details of the earliest known ancestor) is nowhere near enough to do anything useful in mutation studies. See the note I sent earlier this afternoon. John Chandler
16. Re: [DNA] Marker distributions [1]
John W wrote: > One can get a first approximation on the chances of a "random" match by > multiplying the > 12 marker or 25 marker percentages for your set of markers from the tables > John [B] supplied. Emphatically no. This would not even be a "zeroth approximation". There are very strong correlations among the values found at the various loci. That follows from the fact that these are haplotypes -- all the markers are inherited as a unit, so you can't play the game followed in forensic DNA identificat
17. Re: [DNA] STRUNK Y DNA result [1]
Steven wrote: > Using a haplotype arrangement as 388, 393, 392, 19/394, 390, 391, his set > is 12, 13, 11, 17, 25, 10. This is two steps off the Norwegian haplotype > set in Wilson, J. F., Weiss, D. A., Richards, M., Thomas, M. G., Bradman, > N., Goldstein, D. B. "Genetic evidence for different male and female > roles during cultural transitions in the British Isles". Proc. Natl. > Acad. Sci. U. S. A., vol 98 (9) p5078 (2001)". Correction: there is no "Norwegian" haplotype, and the paper by Wilson et al.
18. Re: [DNA] mutations and paper trail [1]
Nel wrote: > haven't had time to take the "tree making" course, assuming I even have the > software to do it. The example shown in my email was composed simply by hand. The secret to making structured material that looks ok regardless of the font used by the viewer is to use dots for making horizontal lines (because they have about the same width as a space in most fonts). John Chandler
19. Re: [DNA] source of data for mutation rates [1]
Ann wrote: > I do recall reading a way to estimate mutation rate by looking at the > distribution of alleles. I think it was an aside in whatever paper I was > reading, and I flagged it to come back to later, but I can't find my source > right now. In principle, it's very simple. You just find a cohort of descendants in the same generation from a common ancestor and tabulate the allele values for all of them. If you calculate the variance of these about the mean and multiply by 4/3, you get the average n
20. Re: [DNA] Kit numbers [1]
I wrote: > to go directly to the graph: > > http://homepages.rootsweb.com/~blanch-l/ftdna.ps Sorry, but I mistyped that. It should have said http://homepages.rootsweb.com/~blanch-l/ftdnakit.ps John Chandler
21. Re: [DNA] Re: Roper website [1]
Bonnie wrote: > I still say it's an awful shame that no one is providing us with data > that really adequately correlate the markers people are getting tested > for, with these detailed phylogenies. Just to inject a dose of reality here -- note that the "detailed" phylogeny of Y haplogroups is just a bare outline of human groups, and is essentially useless for genealogical purposes. You cannot say anything about another member of the same haplogroup besides the fact that you share a common ancestor many t
22. Re: FW: "Surname" groups (was Re: FW: [DNA] Re: Roper website) [1]
Philip wrote: > If I am understanding this correctly, one possible problem with trying to > link SNP's to specific family genealogies is that SNP's mutations may not > be common enough for distinguishing specific family lines. That's not the problem. Don't forget there are billions of base pairs, and most of them are junk DNA. There are sure to be plenty of suitable SNPs. The problem is identifying them and figuring out what group has the mutation and so on. > Or am I wrong and theoretically we could
23. Re: FW: [DNA] Re: Roper website [1]
Bill wrote: > Yes, but we are headed toward more comprehensive testing > thru markers and SNP's that will eventually diverge with the > number of SURNAME projects that being initiated.. Mathematically speaking, "divergence" is the condition wherein some quantity (such as cost) can grow without any limits. Realistically, there are ALWAYS limits. > How many of these SURNAMES could be merged together with > really good DNA testing....until we find maybe 16,000-20,000 > genetic-based SURNAMES.... That's way
24. Re: [DNA] Newspaper account of technical article about Genghis Khan [1]
Ann wrote: > Titillating prose aside, there was one paragraph which I need to think about. > Since it refers to the "average man," a very large percentage of those would > have no male-line descendants living 1000 years later, so I guess that would > bring the average number of descendants down a lot. > > "Early in the last millennium, the population of the world was, speaking very > roughly, 1/20 as large as it is today. Therefore, the average man alive then > has 20 descendants alive today in his direct m
25. Re: "Surname" groups (was Re: FW: [DNA] Re: Roper website) [1]
Phil wrote: > I agree that this may not be a step in the right direction, but would it > really take 16,000 markers for each person? If you test on a few markers, > most people would fall into a few major groups and then only the > appropriate additional markers would need to be tested. That's true, but there is a trade-off of time. If you want to avoid testing the markers that don't actually contribute to specifying your particular surname, you have to start at the top of the phylogenetic tree and work
Viewing 1-25 of 686 matches from 36,222,914 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 0.519921