Results
[I will have to update this section some other time, to reflect the results from our fourth Thrift. Meanwhile, see the Project News section.]
The chart on the "Y-DNA Results" page shows the differences between the DNA results of the five individuals we have in the FTDNA database so far. (In some cases the “most distant ancestor” field is blank; the project member has not filled in this information.) Eventually we will have enough people (with enough markers tested) that we can identify families and group them by their similarities. The current examples illustrate well that DNA testing can in certain cases clearly rule out relatedness. It is harder to clearly prove close relationships; for best results a large number of markers are needed, OR large numbers of families needed to develop a tree showing intermediate relationships and branch points.
The two members of the Thrift family from Virginia illustrate a quirk of statistics: mutations occur completely randomly, and the more markers you test the more accurate your interpretation will be. It is rare but not unheard of to have 3 mutations within 12 markers (that is, 25% mismatched), for fairly close relatives. (If there is a mismatch between descendants then we know that a mutation must have occurred, though we don't know which individual or even which of the two branches it occurred in.) For kit numbers 112459 & 117863, in the first 12 markers there are no mutations, 0% mismatched. In the next 18 markers there are 3 mutations, 17% mismatched; quite a difference. However, we see that even when 67 markers are measured there are still only 3 mutations, giving a rate of <5% mismatched over 67 markers. This is very reasonable for 6th cousins.
Kit #61429 is the same individual as entry 4VZ4N in ySearch.org. The most distant known ancestor is William Andrew Jackson Frith, b. ~ 1834 in Georgia, USA. For the 12 markers tested, there are 2 mismatches for John Frith (at markers DYS390 and 385b), and 3 mismatches for Nathaniel Thrift. As discussed above there is not enough data to determine whether two or even all 3 of these people are closely related, although chances are against it. Although William A.J. Frith appears more closely related to John Frith than to Nathaniel Thrift, more markers need to be tested to establish how close this relationship is.
Between kit #1772, a descendant of Robt T Thrift, and the two kits descended from Nathaniel Thrift there are only 4 markers in common; 8 differences out of 12 markers tested. Because there are so many differences we can safely say these specific families are quite unrelated. (However, we would like to confirm this for all of Robert T Thrift's descendants, by testing a descendant from a different one of his sons. Doing so will prove whether Robert T Thrift possessed the same DNA markers as kit #1772, just as the two descendants of Nathaniel Thrift confirm Nathaniel's DNA to within a few markers.)
Their haplogroup determinations also differ, and this is independent evidence that the ancestors of Nathaniel Thrift and Robt T Thrift are unrelated. Nathaniel Thrift is in haplogroup R1b1 (Iberian peninsula origins, likely precursors of Picts & Celts). Robt T Thrift is in haplogroup I1 (northern European origins, likely Viking, Anglo-Saxon or Norman). Both haplogroups are found in the British Isles, however they arose and were separated in Paleolithic Europe before the last Ice Age -see links below. Thus these two male lines are not related within many thousands of years.
Haplogroup R1b: http://en.wikipedia.org/wiki/Haplogroup_R1b_%28Y-DNA%29
Haplogroup I1: http://en.wikipedia.org/wiki/Haplogroup_I1a_(Y-DNA)
As of January 2009 there is only one THRIFT in the SMGF database (www.smgf.org) (this is Nathaniel, but Jeremiah is listed as the Most Distant Known Ancestor; the family tree at that site doesn’t go back to Nathaniel) and no FRITH or FIRTH. There is one FREED in SMGF. This entry is a 28/37 match for Nathaniel Thrift, not closely related. There is one FRID from Sweden, who matches Nathaniel at only 4 out of 20 markers so is most likely a different haplogroup -as well as a different surname origin.
DNA Test Results Chart:
In the chart on the "Y-DNA Results" page, the column "Haplo" shows the testee's haploGROUP. The columns labeled "393" etc (DYS numbers) are distinct markers. The whole set of markers for a given individual is called that person's haploTYPE. (Haplogroups are different from haplotypes and are tested using different methods, although often the haplogroup can be predicted by knowing the haplotype.) For each marker, an individual has a testable number of repeats (called STR's, short tandem repeats). The number of repeats is indicated in the table. Although these are usually passed down from father to son without change, the repeat numbers at a given marker occasionally do change (mutate), usually by plus or minus 1; these mutations DO happen within historical time, and for each marker the rate at which mutations occur is a constant which is known approximately. (If there were no mutations there would be no point in studying these patterns; everyone would be identical.) The average mutation rate for the STR markers is about 1 per marker per 300 generations; if a whole set of 60 markers is monitored, this predicts roughly one change will be detected per 5 generations, on average. (Remember, the STR's of the Y chromosome are special regions of DNA and are not genes; mutations in actual genes occur by a different mechanism and are much more rare than this.) If two people have 37-marker (or more) haplotypes that are identical, the two people are very likely related (12-marker haplotypes don't provide enough information to be able to say this). If the two people differ by at most 1 or 2 units out of 37 markers or more, they are still likely to be related, but a mutation has occurred in the time since their Most Recent Common Ancestor, thus they are likely to be more distant relatives. Markers shown in red tend to mutate faster than markers shown in black, thus if two people have a 1 unit difference in a red marker, the Most Recent Common Ancestor may be a bit more recent than if there is a 1-unit difference in a black marker.
The wording of the section "DNA Test Results (Alleles) for Project Members" may be a bit confusing; it is set by FTDNA & can't be changed.
