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:mar +date:2003)
Viewing 1-25 of 1,082 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] DYS 464 D = 0 [1]
Grant wrote: > We recently received two 25 marker results from FTDNA with 0 at DYS 464 D. I > searched the archives at genealogy-dna but didn't find a mention of 0 at 464 > D. It happens from time to time. You can see an example in the Rice study (although, for purposes of alignment, we are taking 464c and calling it 464d). > The other scores are ambivalent but the 0 makes me wonder if they related. > > 13 23 14 11 11 14 12 12 12 13 13 29 16 9 10 11 11 25 15 19 28 16 16 18 19 0 > > 13 23 14 11 11 14 12 1
2. Re: [DNA] Percentage of males with same dna results [1]
Malcolm wrote: > If there is say a 150 million American male population then I am interested > to know how many are likely to have a 25/25 match. > > Would some kindhearted soul offer an opinion about my simple calculation. > > It would seem that 2 out of 3 alleles are shared by most males. So 150 > million becomes 100 > > On 390 I am 1 in 9 so 100 million becomes 11 ... No, that's not how to calculate the probability for Y DNA markers, which are all inherited together as a unit, passing from father to so
3. Re: [DNA] Probability question [1]
Roy wrote: > If I have a 50% probability of being related to another person within 7 > generations. And we know for certain that our MRCA can not possibly be within > the first four generations. So those generations are eliminated from the scope. > Then is the probability of being related in the remaining 3 generations > increased? If so, by how much? I'm really glad you asked that question because it reveals a pitfall I imagine many people have fallen into. Answer: yes, a little bit, but not nearly as mu
4. Re: [DNA] confused again [1]
LouAnn wrote: > John H. mentioned bandwidth around the median. I have noticed that Relative > Genetics uses the median as the MRCA on their reports to clients. You mean they report the median as the one-and-only statistic for MRCA? Or do they also report various other confidence levels? > thought we had discussed this and decided that the median number had no > bearing on the probability of MRCA. No, we discussed this, and *some* people decided the maximum likelihood estimate was a poor choice. However,
5. Re: [DNA] Fw: Test Errors [1]
Jim wrote: > To John Chandler: > Where 0.18 came from . . a slight rounding but these are differences in > four pairs where the expected differences are zero and observed differences > are assumed to be error. The averages are simply averaged, Ok, that explains it. I was assuming each comparison was between just two haplotypes. Actually, I'm still puzzled here because, in a group of four haplotypes, there should be six pairs for intercomparison, but a group of three should have only three pairs. How di
6. Re: [DNA] Mutation Rates autosomal STR's [1]
Nancy wrote: > what I thought was really interesting was a noticeable difference in > mutation rates between the same markers in maternal and paternal > meioses. For example, marker FGA (on chromosome 4) had an apparent > mutation rate of 0.01 in maternal meioses and 0.29 in paternal meioses. I'm skeptical. There is a factor that may not have been sufficiently taken into account, namely, false positive results for paternity. In other words, those men may not have been the fathers after all, and so the d
7. Re: [DNA] Please define "two-step mutations" [1]
Gregg wrote: > And as far as the two-step mutations themselves, I will still be interested > to learn the relative probability of a single two-step mutation versus two > single-step mutations. Won't we all! > It seems to me that it could be that some DYS may > have a propensity towards "jumping" values, Yes, that could be, particularly for markers with short repeat sequences (i.e., 2 base pairs). Unfortunately, the only way to assay to be sure that a two-step difference is a single two-step mutation, ra
8. Re: [DNA] King Foobar [1]
Gregg wrote: > John, I am a little confused. If this is your contention, then would you not > also contend that elimination of the first four generations (a la the > original question) from possible MRCA is also not possible? No, I would not. The original question asked me merly to believe that conventional genealogical research can produce facts that are presumably true and that can be tested by DNA analysis (for example). If the tests show that the "facts" don't fit, then I can stop presuming them to b
9. Re: [DNA] 23 out of 25 markers [1]
Grant wrote: > I think it is rare. However, I think a 2 step mutation has been observed and > I think John Chandler said a 3-step mutation is theoretically possible. True, but that's a little stronger than what I actually said. Basically, we have only a general idea of how these mutations occur in the first place. There is no known reason why 3-step mutations could NOT occur. That also goes for 4-step and so on. In fact, there is a mechanism that allows duplication of much larger sequences (as witness t
10. Re: [DNA] Probability question [1]
Gregg wrote: > But I would like to address your question. You are right to note that it is > a senseless artificial construct designed just to check numbers. But it is > not impossible. The point is NOT about whether it is possible to have > different grandparents yet somehow the same parents. The point is about > which generation can possibly be the MOST RECENT common ancestor. Granted, but... > you can imagine a case where, if there is a common ancestor, it would have > had to have happened in the first
11. Re: [DNA] Please define "two-step mutations" [1]
David wrote: > I had assumed that the numbering (23 vs 24 or 25) for a given locus was > arbitrary, perhaps given in chronological order by the discovery date of > the allele, It's interesting you should guess that. When RelGen started out (as Sorenson Genomics), that's exactly how they reported the results. It was a mess! In our first round of testing, the report did not provide any absolute information whatsoever. (The loci were also labeled with arbitrary codes not recognized elsewhere.) What's mor
12. Re: [DNA] Probability question [1]
Bob wrote: > Your elimination of forsensic matches needs some elaboration, since Walsh > seems to have included them in his model. By that, you mean my assumption that the two test subjects are NOT THE SAME MAN. Yes, I assumed that, and I think everybody else did too in this discussion. I deliberately avoided mentioning it, so as not to distract people from the real issues. I suppose a villain or prankster might choose to participate in my DNA study twice under different names, but I'm not going to spen
13. Re: [DNA] Probability question [1]
I seem to have neglected this slant. Gregg wrote: > So back to the question at hand, or a slightly revised one. Let's say you > can exclude generations 2, 3, 4, 5, 6, and 7, but not the first one. I am immensely intrigued by this concept. How can you possibly exclude the 2nd generation without excluding the 1st? If the two testees have a common ancestor in the 1st generation (i.e., if they are brothers), then they also have a common ancestor in the 2nd. If you can show that they have two different pate
14. Re: [DNA] Widening the search window a bit to 23/25 [1]
Jim wrote: > observing mutations - I prefer the word variances as mutations has sort of a > "bad" connotation in the average laymen's thinking, True, but VARIANCE is a technical term used in statistics, so it would cause confusion if you started using it with a different meaning. The best available words are disagreement and difference. > I hear statements such as, "That is one of our fastest moving markers." So, > what does that mean, please? Well, that's an interesting question. As far as I know, th
15. Re: [DNA] How is the distance figured? [1]
Gregg wrote: > The equation used is just the sum of individual differences: > > So let's go through them each - ... > 10th marker; 12 vs 14 = 2 changes ... > 12th marker; 29 vs. 30 = 1 change The answer is right, but the method skipped a step that could have made a difference. Note that the 12th marker is DYS389ii, and that includes all of the 10th marker plus some more. It is therefore necessary to subtract #10 from #12 in both haplotypes before making the comparison. The corrected comparison is: 12t
16. Re: [DNA] Fw: Test Errors [1]
Jim wrote: > This analysis allows > probability of less than 2.5% the two were related. To summarize, out of 23 markers, there were 6 matches, 16 differences of 1 step and 1 difference of 2 steps. Your total of absolute differences is 18. Multiplying that by (1/23)/0.3097 give 2.53. So far, we are on solid ground. I take it that the two numbers in the original table that were expressed to two decimal digits (0.18 and 2.98) were simply mistyped, since neither one could be obtained by adding up integer a
17. Re: [DNA] Probability question [1]
Gregg wrote: > You said you didn't want the ermine King Foobar example, but I am compelled > to now give it anyway: You should have left well enough alone. > King Foobar has two children, Albert and Bob; You're kidding. King Foobar had 7 mistresses and 18 children in all. The same trend to promiscuity is seen in the family all the way down to George and Harold. The number of men carrying the Foobar haplotype is a lot larger than you realized. In fact... > Now come George and Harold to a party where t
18. Re: [DNA] confused again [1]
LouAnn wrote: > For example, the first row across has the following: > > Sample #s; MRCA 10; 95% Confidence Interval Lower Limit 1; Upper Limit 28; > Estimated Year of Birth for Common Ancestor 1815 ... > They say that they use 25 years for one generation and estimate the YOB for > the MRCA using an average participant birth year of 1940. > > What do you think? Calling it an "estimated" YOB is grossly misleading. When you make an estimate, you have to indicate your uncertainty. They need to say either w
19. Re: [DNA] Final Results [1]
Alex wrote: > In addition G. Longoria had a daughter who is alive > and well and living in Texas. These contemporaries of my father were from > another mother and were unknown to my father. You didn't mention whether you mother is living or not. If she is, then you could try an extended "paternity" test that includes you and your siblings and your mother and the living daughter of G. Longoria. If you compare all your siblings' results, you have a good chance of reconstructing a composite double genoty
20. Re: [DNA] confused again [1]
Sue wrote: > I'm still not comfortable with the idea that a mutation might occur every 20 > births when looking at 25-markers. This does not gel even with the very few > results we have - three identical 25-marker tests with men whose probable > MRCA could not have been born prior to 1755...male births in all three lines > from there down total a minimum of - > > (5) line one > (16) line two > (10) line three You're apparently leaving out the matter of relevance. With those wide variations, it's clear tha
21. Re: [DNA] Probability question [1]
There has been a lot of discussion, including analogies and calls for details. I still think the general answer is more instructive, but I can see that the details would help to clear the air. First of all, I don't remember whether the original question specified the lab or the number of markers tested, but we can deduce the number from the fact that the median generation count was 7 -- that means there were about 25 markers. Here is a table, generation by generation, of the a priori probability of findi
22. Re: [DNA] Probability question [1]
I wrote: > No, I can't. It all comes down to the first generation. You say that > the two testees might have the same father or might not. > That means you > are unsure of the identity of at least one of the fathers. That means > you cannot exclude the possibility that the two fathers were, in fact, > brothers. QED. Gregg replied: ] That's right - generation 1, as I said. No, NOT as you said. You said you would permit the MRCA in generation 1, but not in generation 2. If the "two" fathers were actua
23. Re: [DNA] Question about DYS 464e [1]
Susanne wrote: > The Walker DNA Project has their first > participant with a value for DYS 464e. Just as the STR markers occasionally mutate by making an extra copy of the short repeating sequence or dropping a copy, the DNA can make copying errors on a huge scale by duplicating a long sequence or dropping one. We know that this sort of error is especially likely for DYS464 because it has happened three times already (which is why there are normally four copies of DYS464, labeled a-d). You can think of t
24. Re: [DNA] Re: Cohanim Modal Haplotype [1]
Oops. I was caught in the trap of replying to a list message that had a private "reply-to" pointer. Fortunately, Pieter forwarded the message back to me asking if that is what had happened. Herewith the original message. One note added "in proof" -- when I speak of a "clerical" mutation rate, that includes not just copying errors, but all kinds of lab errors... Pieter wrote: > Moereover, mutations can happen at any time, so the rates quoted are best > estimates, not absolute rules. Naturally! The a
25. Re: [DNA] Final Results [1]
Terry wrote: > I was really meaning the final output from the lab tests which are then > translated into the numbers on the reports. I know that the DNA is > amplified then stewed up with various enzymes and markers and things and > I'd supposed it was then analysed by some clever machine or by an > electrophoresis technique. Right on both counts. It's analyzed by a clever machine using electrophoresis. The PCR products are marked with flourescent dyes, and the signal is measured as a function of time in
Viewing 1-25 of 1,082 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 0.539918