Y-Chromosome Markers Families Comparisons

www.roperld.com

Go to Roper male-line Y-chromosome project web page.
Little/Klein/Cline/Kline Genetics Project
Franklin Genetics Project

Because of other time commitments, I will not be analyzing any more Y-markers.

Contents

The results of the three Y-chromosome projects I am running can be seen by the links at http://www.roperld.com/genetics.htm. Several other project that are underway using the services of the company Family Tree DNA can be seen by the links athttp://www.familytreedna.com/surname.html. A list of many Y-chromosome projects is: http://freepages.genealogy.rootsweb.com/~allpoms/genetics1a.html.

The list of 25-markers used to make these calculations

I have used the results for those FTDNA projects that have published the 25-marker results on the Internet or sent them to me to compared them by calculating their "relative mutations",

often called "genetic distance." This is just the sum of the number of relative mutations in the 25 positions.

Relative mutations among the families' 25-markers

Time-Ordered Phylogram

A way to visualize the relationships among the families is by a time-ordered (rooted tree) phylogram calculated from relative-mutations matrices. To create time-ordered phylograms, I use the PHYLIP/Neighbor software using the relative-mutations matrix to generate a tree file (*.tre) to be plotted by the TreeView software. (For a description of how I do it, see PHYLIPTreeViewUse.htm.)

One can use the relative mutations to create a time-ordered (rooted-tree) phylogram, which shows how the families evolved in time:


One can get quite different phylograms by using the Kitsch program instead of Neighbor. I use Neighbor because it gives results in better agreement with the Network program.
I used only the first 21 of the 25 markers for 170 families because of some uncertainties about how to include the four DYS464 markers.
(Only 142 of them are shown in the plot.)
The scale at the bottom left indicates the horizontal distance for 1 mutation from the common ancestor. Vertical distances indicate connections.
From the SNP Y-chromosome notation in http://ycc.biosci.arizona.edu/nomenclature_system/fig1.html for Single-Nucleotide Polymorphisms (SNP), I estimate the nine SNP mutations back to the most recent common ancestor in Africa to be about 6000 years per SNP mutation, using 50,000 years as the time Homo sapiens sapiens left Africa. There are seven mutations back to the F haplogroup's most recent common ancestor (MRCA) for the Roper/Raper testees, or about 39,000 years. Since there are 7 SNP mutations and 14 STR mutations back to the F haplogroup MCRA, at first thought it would appear that the STR mutations per chromosome over long times happen about twice as fast as the SNP mutations. Thus, for long times into the past, each STR mutation is about 2800 years. The time back to the R1b MRCA is about 7 STR mutations times 2800, or 19,600 years, in the last glacial maximum. (Bryan Sykes, on p. 143 in his book, Adam's Curse, a Future Without Men, states that one should not use raw STR mutation rates to try to relate males who are in different SNP haplogroups. This is because, for males so distantly related, some of the markers could have mutated both positive and negative.)
Note that there are two major divisions of the families between Stirling and Boyd3 about a third of the way down. The upper set has 33% of the 170 families in it. The upper set has two major divisions between Tallman2 and Norwalk; 36% of the upper families are in the top set, 12% of the total 170 families. Some major events in and around Europe must have caused these genetic differences between the three sets of families. See below for estimates of the time for the founding fathers of the three groups. If one considers Boyd3 as in error or an outlier family in some other way, the second group down (from Norwalk down to Stirling) is the older group in Europe; that is, it came into Europe first. I call this group G1. The next oldest group in Europe is the bottom group of families, 67% of the 170 families. Although it came into Europe later than G1, it must have come in in larger numbers. I call this group G2. The top group of families came into Europe even later; I call it G3.
So we have:
Group # of Families % of Families
Group 1 30 21%
Group 2 95 67%
Group 3 17 17%
Neolithic time in Western Europe is considered to be since about 7,000 ybp.
Since the G3 (top smallest group in graph) time of arrival in Europe is c3500 ypb, it is definitely a neolithic group. The fact that it is the smallest group also indicates that it is neolithic.
G2 (bottom largest group in graph) would appear to be neolithic; however, if Boyd3 really belongs to this group, it conceivably stretches back to paleolithic times. The fact that there are so many families in this group leads me to believe that this is a paleolithic group and that the sample of families I am using does not include other families more closely related to Boyd3.
G1 (middle group in graph could be considered as paleolithic or right at the paleolithic-neolithic border in time.
Note that G1 (paleolithic/neolithic) and G3 (neolithic) are more closely related than either is to G2 (paleolithic); i.e., they have a common ancestor that has a common ancestor with G2.
The paper The Genetic Legacy of Paleolithic Homo sapiens sapiens in Extant Europeans: A Y Chromosome Perspective by Semino et. al. in Science V. 290, p 1155 Nov 2000 indicates that there were two paleolithic migrations and one neolithic migration into Europe after the Ice Age. As I understand the situation, the two paleolithic groups of people moved into Europe about 30,000 ybp and 22,000 ybp, but had to move out of most of Europe during the Last Glacial Maximum (c18,000 ybp), except for small areas, called refugia, in northern Spain and the Ukraine. They then moved back into Central and Western Europe before the neolithic farmers arrived.

The paper (http://www.genetics.org/cgi/content/full/164/2/781) mentioned above shows that the mutation rates are different for increases and decreases in marker value. The equations for the mutation rates for increasing and decreasing a marker value, taken from that paper, are

The following graph shows the mutation rates for up and down mutations for different marker values:

For the 154 families studied on this web page, the average values of the markers for the Family Tree DNA 25 markers and the calculated mutation rates are:

154 families Y marker 393 390 19(394) 391 385a 385b 426 388 439 389-1 392 (389-2)-(389-1) 458 459a 459b 455 454 447 437 448 449 464a 464b 464c 464d
Average 13 24 14 10 12 15 12 12 12 13 13 16 17 9 10 11 11 25 15 19 30 15 16 17 18
Mutation rate down 2.03E-05 5.62E-04 2.74E-05 8.20E-06 1.50E-05 3.71E-05 1.50E-05 1.50E-05 1.50E-05 2.03E-05 2.03E-05 5.02E-05 6.79E-05 6.06E-06 8.20E-06 1.11E-05 1.11E-05 7.60E-04 3.71E-05 1.24E-04 3.44E-03 3.71E-05 5.02E-05 6.79E-05 9.18E-05
up 4.17E-05 3.77E-04 5.10E-05 2.29E-05 3.42E-05 6.23E-05 3.42E-05 3.42E-05 3.42E-05 4.17E-05 4.17E-05 7.61E-05 9.29E-05 1.88E-05 2.29E-05 2.80E-05 2.80E-05 4.60E-04 6.23E-05 1.39E-04 1.25E-03 6.23E-05 7.61E-05 9.29E-05 1.13E-04

The average mutation rates for the 25 markers used by Family Tree DNA are:

Mutation Direction Mutation Rate
down 0.000221
up 0.000132
average 0.000176

Return to top of page.

If there are enough families in a group of families that has been expanding rapidly from a founder, one can gain some information about that expansion by plotting the frequency of relative mutations; i.e., the number of pairs that have particular relative mutations:

One expects to see a bell-shaped curve if the sample of families is large, random and expanding rapidly. (See The Journey of Man, A Genetic Odessey by Spencer Wells, p.91.) We see two nearly bell-shaped (Gaussian) peaks with peaks at 7.2 ± 2.8 and 24.2 ± 3.9 relative mutations. The first peak is almost entirely due to Group 2 (see below). Group 3 has a very small peak hidden under the first peak (see below). The small peak for Group 1 (see below), the oldest group, is hidden under the second peak.
The fact that the location of the second peak is larger than the first peak implies that the relative mutations between Group 2 and the other two Groups indicates a common ancestor at the peak value. One can improve the fit to the data by including another Gaussian peak between these two peaks, which corresponds with a peak for Group 1 shown next.
The following three graphs show the frequency of mutations for the three groups.

More data are needed for families in this group to make the curve smoother.
The peak appears to be at about 20.7 ± 4.2 relative mutations.

The peak is at 7.2 ± 2.7 relative mutations.

More data are needed for families in this group to make the curve smoother.
The peak appears to be at about 8.7 ± 3.1 relative mutations.
This curve is hidden under the much larger curve for Group 2 in the graph above for all families.

Return to top of page.

One can compare the medians for markers for the three groups:

Marker Name
(prefix DYS)		 393 390 19(394) 391 385a 385b 426 388 439 389-1 392 (389-2)- (389-1) 458 459a 459b 455 454 447 437 448 449 464a 464b 464c 464d
Group 1 14 23 15 10 15 16 11 13 11 13 11 17 16 8 9 11 11 25 14 20 30 13 15 16 16
Group 2 13 24 14 11 11 14 12 12 12 13 13 16 17 9 10 11 11 25 15 19 29 16 16 18 18
Group 3 13 23 14 10 14 14 11 14 11 12 11 16 15 8 9 8 11 23 16 20 29 13 15 16 17
Maximum-
Minimum		 1 1 1 1 4 2 1 2 1 1 2 1 2 1 1 3 0 2 2 1 1 3 1 2 2

Note that the three markers with the greatest differences are DYS385a (4), DYS455 (3) and DYS464a (3). When data for a new family becomes available, I use these three markers to decide to which of the three families it belongs. Of course, one can also do a relative-mutations calculation with the markers for a new family and the three groups.

The relative mutations for these groups are:

Relative
Mutations		 Group 1 Group 2 Group 3
Group 1 0
Group 2 22 0
Group 3 17 23 0

So, Group 1 and Group 3 (17 relative mutation) are more closely related than are Group 1 and Group 2 (22 relative mutations) or Group 2 and Group 3 (23 relative mutations).

Return to top of page.

Haplogroup assignments

Haplogroup determinations

Family Tree DNA can now determine the haplogroup for a male according to the latest nomenclature for haplogroups. I have had mine done and some others have informed me of theirs. Here are the ones I know so far:

Markers (prefix DYS)
Name Haplogroup 393 390 19(394) 391 385a 385b 426 388 439 389a 392 389b 458 459a 459b 455 454 447 437 448 449 464a 464b 464c 464d
Roper (LDR) R1b 13 24 14 10 11 16 12 12 12 13 13 17 16 9 10 11 11 25 15 19 30 15 16 16 17
Talbot E3a 14 22 15 11 17 18 11 12 13 13 11 17 17 8 10 11 11 25 14 21 29 14 17 18 18
Stidham E3b 13 24 13 10 16 18 11 12 12 13 11 16 15 9 9 11 11 25 14 21 31 15 17 18 18

The paper http://evolutsioon.ut.ee/publications/Tambets2004.pdf gives some Y-haplogroup percentages for some regions and populations(p.7).

Comparison of families with 37 markers

The following table shows the relative mutations for some families for which 37 markers have been measured:

Relative
Mutations		 37
Markers		 LDR WAR SR IBF CMF JEF BF JMAL TNL Davnpt9415 Davnpt8666 Davnpt7181 Davnpt7505 Davnpt11950 Davnpt11382 Davnpt11596 Sutton4933 Sutton4141 Duerinck281 Arnold6078 Ivey7724 Ivey10521 Ivy10541 Ivie8367 Ivy13261 Ivey7362 Ivy8355 Ivey10508 Hurst10507 Hurst1370 Hurst3131 Hurst8764 Hurst13991 Bricker
1 LDR 0 LDR
2 WAR 1 0 WAR
3 SR 12 11 0 SR
4 IBF 15 16 12 0 IBF
5 CMF 14 15 11 1 0 CMF
6 JEF 13 14 12 2 1 0 JEF
7 BF 15 16 12 2 1 2 0 BF
8 JMAL 7 8 15 16 15 14 14 0 JMAL
9 TNL 34 35 34 32 31 32 30 33 0 TNL
10 Dv9415 14 13 16 12 11 10 12 15 31 0 Dv9415
11 Dv8666 16 15 16 14 13 12 14 17 33 2 0 Dv8666
12 Dv7181 15 14 17 11 10 11 11 16 32 3 3 0 Dv7181
13 Dv7505 15 14 17 11 10 11 11 16 32 3 3 0 0 Dv7505
14 Dv11950 14 13 16 10 9 10 10 15 31 3 5 2 2 0 Dv11950
15 Dv11382 16 15 20 14 13 12 14 17 35 5 5 4 4 4 0 Dv11382
16 Dv11596 19 18 15 17 16 17 17 16 38 15 17 16 16 16 20 0 Dv11596
17 Sut4933 46 47 45 46 45 46 46 45 45 45 45 46 46 45 47 48 0 Sut4933
18 Sut4141 46 47 45 46 45 46 46 45 45 45 45 46 46 45 47 48 0 0 Sut4141
19 Drk281 11 12 16 13 12 11 13 16 38 15 17 16 16 16 18 18 50 50 0 Drk281
20 Ard6078 14 15 18 14 13 14 14 13 30 17 19 16 16 15 19 16 43 43 16 0 Ard6078
21 Ivey7724 39 40 45 48 47 48 48 46 46 44 44 43 43 43 45 49 42 42 41 39 0 Ivey7724
22 Ivey10521 39 40 45 48 47 48 48 46 46 44 44 43 43 43 45 49 42 42 41 39 0 0 Ivey10521
23 Ivy10541 40 41 46 49 48 49 49 47 47 45 45 44 44 44 46 50 43 43 42 40 1 1 0 Ivy10541
24 Ivie8367 39 40 45 46 45 46 46 46 46 44 44 43 43 43 45 49 42 42 39 37 2 2 3 0 Ivie8367
25 Ivy13261 38 39 44 47 46 47 47 45 45 43 43 42 42 42 44 48 41 41 40 38 1 1 2 3 0 Ivy13261
26 Ivey7362 38 39 44 47 46 47 47 45 45 43 43 42 42 42 44 48 43 43 40 40 3 3 2 5 2 0 Ivey7362
27 Ivy8355 39 40 45 48 47 48 48 46 44 44 44 43 43 43 45 49 42 42 41 39 4 4 3 6 3 3 0 Ivy8355
28 Ivey10508 30 31 33 32 31 32 30 31 20 29 31 30 30 29 33 36 47 47 32 28 40 40 39 40 39 37 38 0 Ivey10508
29 Hurst10507 17 18 22 21 20 21 21 16 38 21 21 20 20 20 24 20 42 42 20 18 43 43 44 43 42 44 43 38 0 Hurst10507
30 Hurst1370 14 15 19 18 17 18 18 13 35 18 18 17 17 17 21 17 39 39 17 15 40 40 41 40 39 41 40 35 3 0 Hurst1370
31 Hurst3131 14 15 19 18 17 18 18 13 35 18 18 17 17 17 21 17 39 39 17 15 40 40 41 40 39 41 40 35 3 0 0 Hurst3131
32 Hurst8764 14 15 19 18 17 18 18 13 35 18 18 17 17 17 21 17 39 39 17 15 40 40 41 40 39 41 40 35 3 0 0 0 Hurst8764
33 Hurst13991 13 14 18 19 18 19 19 14 36 19 19 18 18 18 22 18 40 40 16 16 39 39 40 39 38 40 39 36 4 1 1 1 0 Hurst13991
34 Bricker 15 16 22 10 11 12 12 18 35 15 17 14 14 13 17 24 44 44 17 19 48 48 47 46 47 45 46 30 21 20 20 20 21 0 Bricker
LDR WAR SR IBF CMF JEF BF JMAL TNL Dv9415 Dv8666 Dv7181 Dv7505 Dv11950 Dv11382 Dv11596 Sut4933 Sut4141 Drk281 Ard6078 Ivey7724 Ivey10521 Ivy10541 Ivie8367 Ivy13261 Ivey7362 Ivy8355 Ivey10508 Hurst10507 Hurst1370 Hurst3131 Hurst8764 Hurst13991 Bricker

References:

This is a phylogram for the families with 37 markers:

Homo sapiens sapiens events

I have data for enough families now to see the general pattern of European families.

Because of other time commitments, I will not be analyzing any more Y-markers.

Go to Roper male-line Y-chromosome project web page.
Little/Klein/Cline Genetics Project
Franklin Genetics Project

Return to top of page.
www.roperld.com