As mentioned in the Table I discussion, a single haplotype (the set of marker values for a participant) is meaningless, i.e., it is just a set of numbers. It is only when one haplotype is compared with one or more others that meaning begins to emerge. Such comparisons address the relationships between and among participants, and hence lineages. Those relationships are the subject of Table II.
Table II presents the data in Table I is a different manner. A close examination of the marker values in Table I shows that some participants have haplotypes very close to, and sometimes identical to, other participants. Others are quite different. These similarities and differences have led us to develop 34 family groups, labeled Group 01 - Group 34. Additional testing may lead to the creation of additional family groups.
Family Groups
For each group, a reference row is shown at the top. This row contains an arbitrary set of marker values that seems to represent the majority of values of the participants in that group. Then, for each participant each marker value is subtracted from the reference value and the difference is entered in Table II for that participant. Some participants have no difference from the reference in 25 markers (all dashes). Others have a deviation of one, or two, or three, or, in one case, four from the reference in 25 markers. The deviations in Markers 26-67 are somewhat greater. However, for purposes of these "family groupings," only the values in Markers 1-25 were used. Changes (mutations) from the reference are highlighted.
The three columns on the right of Table II show the total deviations (or genetic distance) from the reference values at Markers 1-25, Markers 1-37, and Markers 1-67. This genetic distance is just the number of mutations with a small adjustment for special considerations.
To compare the reference haplotypes, see Table III. That table shows how far these groups of participants are from one another, even though they are all named Beatty (various spellings).
Change of Marker Values (Mutations)
Geneticists, using complex statistical models, tell us that on average a marker changes about once in 500 generations or in about 12,500 years (assuming about 25 years per generation).
So, it takes about 20 generations to make a change in a 25-marker haplotype (500 generations divided by 25 markers). Said another way, within 20 generations, ONE (and only one) of the 25 markers is likely to change from one value to another (usually either up one or down one). That amounts to one change in a 25-marker haplotype in 500 years! For a 25-marker haplotype, the son is identical to the father approximately 95% of the time.
It is important to keep in mind that these changes are not according to any pattern; they just happen randomly as sperm cells are being created. Once a son receives a modified Y-chromosome that modification is passed on to his descendants, providing a testable label.
Because of the slow rate of change, the test results are most valuable as indicators of rather distant relationships. It is essential to look at the results from a statistical point of view. An important concept is that of the most recent common ancestor (MRCA) of two men. The time to MRCA is naturally expressed in generations, but one can convert to years by assuming a typical generation length (most geneticists use 25 years per generation). The analysis below is intended to reach some conclusions based on the test results with some general genealogical and historical patterns.
Please be aware that although the rates of change described above are useful for our purposes, they are gross generalizations. For one thing, it has been determined that not all markers mutate at the same rate. Some are considered "fast movers." They tend to mutate more rapidly than others. Presumably, there could also be "slow movers" as well, although they probably haven't been identified as such.
Another consideration is the vagaries of statistics. To illustrate, considerate that when you flip a coin, there is a 50% statistical chance that it will come up Heads and a 50% statistical chance that it will come up Tails. However, if you took up a coin right now and flipped it three times, you might get three Heads in a row. Does that invalidate the statistics? Not at all. If you continued to flip the coin another 97 times, you would likely get 50 Heads and 50 Tails. And, 1,000 flips would get close to 500 Heads and 500 Tails. So, it is possible (though VERY unlikely) that a 25-marker haplotype could mutate in each of three succeeding generations (rather than once in 20 generations).
Who Is Related and Who Isn't?
As mentioned elsewhere, a man's haplotype is useful only for comparing with those of other men to get a clue to how closely they are related. Relatively closely related men have similar haplotypes and more distant relationships are indicated by larger differences. A genetic distance of 0 in 25 markers (a perfect match) means that about half the time the two men will have an MRCA within seven generations. That implies that half the time the MRCA will be more than seven generations away, with the possibility of it being many generations.
An explanation of "not related" is needed. If the MRCA for two men is earlier than the era of adoption of inherited surnames, there is not much prospect for discovering a family relationship, and it is reasonable to say they are unrelated. The era of surname adoption in Britain was about 1200-1400. The rule of thumb for these Y chromosome tests using 25 markers is that if the genetic distance is four or more the men should be considered unrelated unless there is some other indicator of a relationship.
Remember, we are using DNA test results as an adjunct to traditional genealogy research. Hence, the usefulness of this data is greatest when applied to the more 'recent' past (say, within the past 500-700 years). This is where the 'related' and 'unrelated' categories are helpful. We do NOT contend that the 'unrelated' groups shown in Table II did not arise from one or more common ancestors hundreds (perhaps thousands) of generations ago. In fact, there are scholarly arguments that suggest that they did. However, those suppositions and possibilities are beyond the scope of this project.
Grouping The Results
Group 01
The groups are determined by simply looking for results that are close. In the case of Group 01, 24 haplotypes (25-marker) are identical and a genetic distance of only one separates another 48 from them. This strong clustering indicates a common ancestor with the reference haplotype. As an aid to discussion we have named this common ancestor "Adam." (Of course, this offends the kinfolks in the other groups who feel like they each have their own "Adam," which they do. Oh well. In this family, majority rules.)
Group 01 contains about 69% of the men tested thus far. Most of those have haplotypes which differ from the indicated common ancestor, Adam, with a genetic distance less than 3 at 25 markers. A few others appear to be related to some of these men and thus are also descendants of Adam. This suggests that these closely related men fit into a family group with a common ancestor living about 20-30 generations ago. The indications are that Adam lived before the year 1600 (perhaps before 1400), and because he probably lived after the adoption of hereditary surnames, his surname was probably Beatty (or a variant thereof). We will probably never know much more about this Adam. Knowledge that Group 01 has a common ancestor and approximately where and when he lived allows us to try to follow the paths whereby the ancestral lines went from Adam to the present.
Based on the limited information available, an attempt has been made to graphically represent what the Adam Lineage might look like. The lineage chart if highly speculative and should not be considered as authoritative. To see the chart, click here.
Group 02
Group 02 contains two men in L-44 (ID-4 and ID-18). These men are known to be first cousins (they have the same paternal grandfather) so it is not surprising that their haplotypes are identical. Their haplotypes, however, are very distant from all other men tested in this project.
A third member of L-44 (ID-35) has also been tested. His results were quite surprising. He is definitely not related genetically to the two men in Group 02. He is close to ID-115 (L-124) and ID-30 (L-142), and they appear to be descendants of Adam although the distance from Adam is large enough to cause concern. The genealogical record says that the three men of L-44 share a common ancestor in Thomas Beaty of Cross Creek, PA who died in 1816.
Inasmuch as ID-35 is a descendant of Adam it is reasonable to regard the men in Group 02 as cultural descendants of Adam except that somewhere along the way a child was given the Beatty name even though he had a Y chromosome from another man.
Recently, another surname DNA project has shed some revealing light on the genetic ancestry of ID-4 and ID-18. Andrew Lancaster, administrator of the Lancaster DNA Project, has created a compilation and analysis of genetic data for several participants in that project who share a distinctive mutation pattern. Over a dozen men have repeat numbers of 25, 10, and 13 for markers 390, 391 and 392, respectively, and numbers 15 and 15 at markers 385a and b. Andrew has dubbed this group the MacWhos. Surnames in this group include Beatty, Porter, McLane, Furry, Fergie, Fergus, Ferguson, McClay, Wilson, and Livingstone.
ID-4 and ID-18 are exact matches at 25 markers to a man named Porter and one named McLane. In addition, ID-4, but not ID-18, is an exact to Porter at 37 markers.
All members of this group appear to have Scottish origins, although several trace their ancestry through Northern Ireland before coming to America.
A more detailed discussion of this group can be found at:
http://users.skynet.be/lancaster/MacWho_results.htm
Also, ID-4 has written an interesting description of this issue which can be read by clicking here.
The information on the Lancaster site clearly points to a non-paternity event. The three cousins of L-44 have compared notes and identified a time interval when the Porter genes came into that part of Lineage 44 represented by ID-4 and ID-18. That interval was the generations between Thomas Beaty, born 1747 in Ireland, immigrated to North America by the 1770s, and died 1816 in Pennsylvania USA, and Walker Beatty (his great grandson), born 1854 in Ohio USA and died 1932 in Iowa USA. The haplotype of Thomas Beaty appears to be part of Group 01. However, the haplotype of his great grandson, Walker Beatty, is part of Group 02. So it's most likely that the non-Beaty paternity or unregistered adoption occurred in the USA.
Group 03
Group 03 has had a major genealogical breakthrough in the past few months. There are now three members in Group 03 from different BP-2000 lineages - L-262, L-199, and L-351. The three men are genetically close to one another but very different from all other men tested. Previously a genealogical connection had not even been suspected. The indication is that the three men tested have a common ancestor in the not too distant past and that two of the lines of descent had a subsequent mutation at DYS-449 (Marker 21) and at DYS-456 (Marker 30), respectively. There is no clue concerning which lines had the mutations. We have arbitrarily selected ID-66's haplotype as the Reference haplotype of Group 03.
As previously mentioned, some of the markers have been recognized as "fast moving," meaning that the mutation rates are higher than the others. These markers designated as fast have their names displayed in red in Tables I and II. In the case of Group 03 the difference between two of the haplotypes is at one of the fast markers (Marker 21), meaning that the genetic distance is not as great as it might otherwise seem. The most recent common ancestor of Group 03 is likely to be within a genealogically relevant time frame. Members of L-262, L-199, and L-351 are searching for the connection.
Groups 06 and 15
Groups 06 and 15 are two more examples where different lineages were found to be closely related. Each group contains two men. Within each group, the men are separated by a single mutation. In Group 06, the mutation occurred in DYS-439 (Marker 9). In Group 15, the mutation occurred in DYS-459a (Marker 14). It is interesting to note that DYS-439 is considered a "fast mover." Again, both groups are quite different from all other groups as well as each other.
Group 08
Until recently, Group 08, representing Lineage 2B, contained two men (ID-46 and ID-49) whose haplotypes were different by a single mutation. Recently, the 25-marker test results of another participant (ID-99) were posted. The haplotype of ID-99 is an exact 25-marker match to ID-49. Such information can provide important clues to the ancestry of all three men and illustrates the value of selective sampling.
Groups 10, 12 and 21
These three groups are similar to one another in that each contains two men of the same lineage. As would be expected, the two men in Group 12 (ID-59 and ID-64) have identical haplotypes. However, the two men in Groups 10 and 21 are separated by a single mutation within 25 markers. (Group 10 shows additional differences at Markers 33 and 34.) It is clear that the pairs of men in each group are closely related although the three groups are unrelated.
Group 14
This group contains two participants, ID-71 and ID-105. FamilyTreeDNA (FTDNA) considers the genetic distance between them to be 2 at 25 markers.
It is a bit strange because they agree exactly at all markers except DYS464. DYS464 has some quirky characteristics. It has 4 copies in almost everyone, but with some men it is difficult to count. The results are reported in ascending order. At this set of markers ID-105 agrees with the R1b mode at 15-15-17-17, whereas ID-71 has 15-15-15-15. That suggests that ID-105 and ID-71 are different by gd=4, but FTDNA has a special way of counting. With the 464 marker, for any of the copies, a change of any size is counted as 1. This is part of the quirkiness. With these two at gd=2, it is ID-71 who has the unusual haplotype.
When compared with the whole FTDNA database neither man has any matches which FTDNA considers significant, except with each other. It appears that these two are way out in right field by themselves. The simple interpretation is that they have a common ancestor, not so long ago, with ID-71 having received whatever mutation(s) produced the 15-15-15-15 later.
The conversion of 15-15-17-17 to 15-15-15-15 is likely a special class of mutation called recLOH. The recLOH mutations are distinctly complex, involving structures called palindromes. The main effect is copying one part of a multi-copy marker onto another part, possibly causing a large change in one event. A second effect is that more than one marker can be changed in one event. If indeed ID-71's unusual result is a recLOH, it means that ID-105 and ID-71 are separated by only one mutation.
For these two participants, the test results provide an interesting situation which opens a number of doors for further research.
Group 24
This group contains three participants of apparently different lineages. However, at 25 markers, two carry exactly the same haplotype suggesting that those two are closely related. They do differ at two markers in the 26-37 marker panel. Still, the relationship appears sufficiently close to warrant further cooperative study between these two lineages.
The Remaining Groups
Twenty one groups have only one member so far, making comparisons within the groups impossible. What is clear from the tests is that they are not related to each other or to any of the other men tested. No doubt other men will fall into these groups; they just haven't been found.
The Interesting Case of Lineage 2
The test results shed some interesting light on BP-2000's L-2. This is a well documented lineage and was one of the first documented in BP-2000. However, as DNA test results began to come in, it became clear that additional work was needed on this lineage. Thus far, 18 men in L-2 have been tested. Of these, four are clearly not related to the other 14. One is placed in Group 11 and the other three in Group 8. The remaining 14 are placed in Group 01, but the differences among them have caused the L-2 researchers to look again at relationships within L-2.
As with the case of Group 02 discussed above, the men in Groups 08 and 11 are still considered culturally as part of L-2. The test results suggest that their Y-chromosomes came from a male outside the genetic L-2 lineage.
What About Markers 26 - 37?
One hundred and eighteen of the 160 men tested have upgraded to 37 markers. The extra data were not included in the above discussion because the results are new to the project and interpretation is unfamiliar. Actually these tests are rather new to the lab as well.
The values at two of the markers, CDYa/b (Markers 34 and 35) appear to be special cases. In Group 01, it is unclear what Adam's allele is at CDYa. In spite of these problems some interesting conclusions can be reached. Twenty Two of the 24 men who carried Adam's haplotype at 25 markers have upgraded to 37 markers. The extra data showed a division based on DYS-456 (Marker 30). The values at CDYa tend to reinforce the division, however this is subject to a reinterpretation.
More data and more experience with the extra 12 markers should help with the interpretation.