Hogg DNA Project R1b cluster no. 1


This cluster currently consists of 23 members. Five of them have the surname Hoge/Hogue and are confirmed or suspected descendants of William Hoge and Barbara Hume, who came to America on the ship CALIDONIA ariving at Perth Amboy, New Jersey, in about 1682 (the line identified as NJ1682). In addition, there are several other individuals with matching DNA having other surnames including Carr/Kerr/Karr/Cerr, Hamilton, McLean, McIntire, Kilgour, Roberts, and Middleton. The DNA match indicates that all of these people share a common ancestor.

Minimal pedigrees for the Hoge/Hogue members of the cluster that we can trace back to William Hoge and Barbara Hume are shown here:

   1 Sir John Hoge of Musselburgh Scotland, b. before 1575
     2 Sir George Hoge of Musselburgh Scotland, b. before 1613
       3 Sir James Hoge of Musselburgh Scotland, b.1609-1638, d.1682
         4 William Hoge b.1660, Musselburgh Scotland, d.1745, Frederick Co. VA
           5 William Hoge Jr. b.1708, Perth Amboy NJ, d.21_Apr_1759, Loudon Co. VA
             6 James Hoge b.6_Dec_1724, Chester Co. PA
             . 7 Jacob Hoge b.1755, d.1817
             .   8 James Hogue b.ca.1780, Dallas, Paulding Co. GA
             .     9 William S. Hogue b.1799, Paulding Co. GA d.26_Feb_1883, Paulding Co. GA
             .     .10 William Jacob Hogue b.9_May_1830, Paulding Co. GA d.19_Jan_1905, Paulding Co. GA
             .     .  11 William Thomas Hogue b.May_1853, Georgia
             .     .    12 Dr. William L. Hogue b.Sep_1880, Georgia d. before 2001
             .     .      13 Dr. William Leroy Hogue b.1915, Draketown, Haralson Co. GA d.29_Jul_2001, Augusta GA
             .     .        14 Project NJ1682c: desc of William Hoge > William > James #2
             .     9 Jacob Hogue b.4_May_1800, Dallas, Paulding Co. GA d.7_May_1873, Marshall, Harrison Co. TX
             .      10 Jonathan Patrick Hogue b.ca.1847, Draketown, Haralson Co. GA, d.1895, Bright Star AR
             .        11 Ben Dee Hogue b. 29_Mar_1879, Arkansas
             .          12 Clifford Easters Hogue b.1_Jun_1918, Texas
             .            13 Project ID NJ1682a: desc of William Hoge > William > James #1
             6 George Hoge b.6_Feb_1732/33, Frederick Co. VA d.1805, Green Co. PA
               7 Thomas Hoge b.25_Sep_1765, Green Co. PA, d.4_Jan_1837, Green Co. PA
               . 8 Abner Hoge b.1_Nov_1812, Green Co. PA, d.26_Feb_1890, Wenono IL
               .   9 Thomas Jefferson Hoge b.10_Apr_1840, Green Co. PA, d.22_Jan_1905, Wenona IL
               .    10 John C. Hoge b.10_Sep_1887, Wenona IL, d.20_Feb_1960, Phoenix AZ
               .      11 Project ID NJ1682b: desc of William Hoge > William > George, b.1732
               7 ________ Hoge (this connection is a guess)
                 8 Isaac Hoge of Greene Co. PA b.1801, Pennsylvania
                   9 Edward Hoge b.1834, Pennsylvania d.1910, Pennsylvania
                    10 Isaac F. Hoge b.1870, Ohio d.1962, West Virginia
                      11 Russell Hogue b.1902, West Virginia d.1958, Arizona
                        12 Denver Hogue b.1923, West Virginia d.1987, Georgia
                          13 Project ID NJ1682d: desc of Isaac Hoge of Greene Co. PA

Note that all of these are descendants of William and Barbara's son William Hoge Jr. We would like to have data from descendanta of two of William and Barbar's other sons. With data from three sons we can infer the haplotype of William, the ancestor of all of them. With that, we will be better suited to understand the relationship between the Hoge/Hogue's and the other members of the cluster because we can eliminate any confusion that might arrise due to mutations more recent than William. The other Hogue in the cluster is a descendant of Montgomery Hogue, b.1820, Georgia (GA1820a). We presume that he is also a descendant of William and Barbara, but we cannot determine how he is connected. The minimal pedigree for him is shown here:

   1 Montgomery Hogue b.1820, Georgia
     2 Thomas Jefferson Hogue b.29_Sep_1855, Holly Springs, Dallas Co. AR
       3 Emmitt Leonard Hogue b.27_June_1896, Falls Co. TX d.17_Nov_1978, Howard Co. TX
         4 Bruce Leonard Hogue b.1925, Texas
           5 Hogg DNA Project ID GA1820a

Other than their different surnames, and in some cases their earliest known ancestor, we don't have as much information about the other members of the cluster. Most of them descend from an ancestor that emmigrated to America, offering the possibilty that their connection to the Hoge/Hogue line might be here (in America) during the colonial period. However, one that we know of, Mr. Hamilton, descends from ancestors that did not emmigrate to America, indicating that his connection to the Hoge/Hogue line is in Scotland before 1682. There are some members of the cluster for which all we know is their surname and their DNA data.

The following table is a brief compilation of the DNA data from all members of the cluster. Here we have included only the data for markers for which one or more members are different than the modal; i.e. markers for which all members have the same value are not shown. We also show the estimated mutation rate (generations per mutation) for each marker. We have developed a model for the tree descending from the common ancestor, shown below. The assumed haplotype of the common ancestor is also shown in the table and any value for an individual that differs from the assumed ancestral haplotype is flagged with a pair of astrices. The mutation rates shown in the table are from http://freepages.genealogy.rootsweb.ancestry.com/~geneticgenealogy/ratestuff.htm

                   DYS  DYS  DYS  DYS     DYS464  DYS  GAT  DYS            DYS  DYS  DYS  DYS
marker             390  458 359b  448    a/b/c/d  460  AH4  570 CDYa CSYb  442  534  461  439

mutation rate
(generations       322  123  442  741      177    249  481  127   28   28  309  120  430  210
 per mutation)

assumed
ancestral           24   16    9   20   15-17-18   11   11   18   36   38   12   15   12   13
haplotype

proj ID  surname    data... (*xx* indicates a mutation)

modal06             24   16    9   20   15-17-18   11   11   18   36   38   12   15   12   13
GA1820a  Hogue      24   16    -   20   15-17-18   11   11   --   --   --   --   --   12   13
NJ1682a  Hogue      24   16    9   20   15-17-18   11   11   18   36   38   12   --   --   13
NJ1682b  Hoge       24   --    -   --   --------   --   --   --   --   --   --   --   --   13
NJ1682c  Hogue      24   16    9   20   15-17-18   11   11   18   36   38   12   15   --   13
NJ1682d  Hogue     *23*  16    9   20   15-17-18   11   11   18   36   38   12   --   --   13
other06  Hamilton   24   16    9   20   15-17-18   11  *12*  18   36  *37*  12  *14*  --   13
other07  Carr       24   16    9   20   15-17-18   11   11   18  *37*  38   12  *16*  --   13
other08  McLean     24   16    9   20   15-17-18  *10*  11   18   36  *37*  12   --   --   13
other09  Carr       24   16  *10*  20  *15-17*     --   --   --   --   --   --   --   --   13
other10  Kerr       24   16    9   20  *15*        --   --   --   --   --   --   --   --   13
other11  Cerr       24   16  *10*  20  *15-16-17*  11   11   --   --   --   12   --   --   13
other12  McIntire   24   16    9  *19* *15-17*     11   11   --   --   --   12   --   --   13
other13  Carr       24   16    9   --   15-17-18   --   11   --   --   --   12   --   12   13
other14  Karr       24   16    -   20  *15-17*     11   11   --   --   --   --   --   12   13
other15  Carr       24   16    -   20  *15-17*     11   11   --   --   --   --   --   12  *11*
other16  Carr       24  *17*   -   20  *15-17*     11   11   --   --   --   --   --  *13*  13
other23  Kilgour    24   16  *10*  20  *15-17*     11   11   18   36  *36*  12  *14*  --   13
other24  Roberts    24   16  *10*  20  *15-17*     11   11  *19*  36  *37* *11* *14*  --   13
other28  Middleton  24   16  *10*  20  *15-17*     11   11   --   --   --   12   --  *13*  13

The goal at this point is to use the information in hand (DNA data, surnames, and pedigrees) to infer the tree describing the descent of each cluster member from the unknown common ancestor. We do not persume to know the surname of the unknown common ancestor, or for that matter, that he had a surname. It is possible, perhaps likely, that the common ancestor lived in medieval Scotland prior to the selection/assignment of surnames. However, some of the surnames in the cluster were in use very early, so one of them might have been used by the common ancestor.

The trees to follow start with the unknown common ancestor on the left and, working to the right, branch as necessary to reflect a DNA mutation or a surname change. In principle, there are a very large number of possibilities and finding the correct tree with absolute certainty is not possible. Our approach is to start with logic and reason and use the science and mathematics of probability and statistics to infer our best guess at the correct tree. DNA mutations are random events and anything can happen in any given case, however, since DNA mutations are rare events, some possibilites have very high probability and others have very low probability. In the jargon of mathematics we are using a "maximum likelihood" approach. We will consider all reasonable possibilities and choose the one with the highest likelihood of being correct. But, it is important to remember that unlikely events do sometimes happen: It would have been considered unlikely that the TITANIC would strike an iceburg and sink on her maiden voyage.

Given complete knowledge of the probabilities of mutations in each marker and the willingness to actually consider every concievable possibility, we could construst a rigorous maximum-likelihood selection of the most likely tree. However, since mutations are very rare events, we can simplify the process. In practice, finding the maximum-likelihood answer is equivlant to finding the one that has the smallest number of mutations. Counting the number of different values at each marker in the table above and adding them up, we find that there must be at least 18 DNA mutations. If we could find a tree with only 18 mutations, that tree would be much more likely than any other tree requiring more than 18 mutations. If we were able to find more than one tree with only 18 mutations, we would not be able to choose any one over the others without doing the rigorous calculation based on specific probabilities of indivual markers, and even then, the differences in likelihood will be modest at best.

It is usually possible to find one marker that appears to separate the entire set into two groups. That would be the earliest mutation closest to the trunk of the tree. In this case, that appears to be either DYS359b, which is 9 or 10, or DYS464, which is 15-17-18 or 15-17. However, we cannot tell if the unknowh common ancestor had DYS359b=9 and DYS464=15-17-18 or if he had DYS359b=10 and DYS464=15-17. We have no choice but to try both possibilities. In practice, by assuming one or the other of these two possibilities, we are making an assumption of which of the two domonant groups, the Hoge's with DYS359b=9 and DYS464=15-17-18, or the Carr's with DYS359b=10 and DYS464=15-17, are on the trunk of the trees. We will construct two trees based on these two possibilities. If one of them has fewer mutations than the other, that will be the one most likely to be correct.

The trees that follows begin with the unknown common ancestor on the left and branch out to the right. Beneath the name "unknown name" are the values assumed for him for all of the DNA markers that show a later mutation. Under each name on the branches to right are the values for the markers that, for that branch, differ from what was assumed for the unknown common ancestor. In each case, when the value for a given marker differs from the value of the preceeding branch, indicating a mutation, that marker is flagged with an asterisk.

Tree no. 1: This is the tree that results from assuming the unknown common ancestor had DYS359b=9 and DYS464=15-17-18. Since most of the Hoge's have these values, this tree has the Hoge's on the trunk and the Carr's as a branch.

unknown name ---+- Hoge/Hogue ---+- Hoge/Hogue - NJ1682a/b/c, GA1820a 390=24 | | 458=16 | +- Hogue - NJ1682d 359b=9 | *390=23 448=20 | 464=15-17-18 +- Haml./McLean -+- McLean - other08 460=11 | *CDYb=37 | CDYb=37 GATAH4=11 | | *460=10 570=18 | | CDYa=36 | +- Hamilton - other06 CDYb=38 | CDYb=37 442=12 | *GATAH4=12 534=15 | *534=14 461=12 | 439=13 +- Carr et. al. -+- Carr - other13 | +- Carr - other07 | *CDYa=37 | *534=16 | +- Carr et.al. --+- Carr et.al. --+- Carr/Karr ----+- Carr/Karr - other09/14 *464=15-17 | 464=15-17 | 464=15-17 | 464=15-17 | *359b=10 | 359b=10 | 359b=10 | | | | | +- Carr - other15 | | 464=15-17 | | 359b=10 | | *439=11 | | | +- Carr/Middl. --+- Middleton - other28 | | 464=15-17 | 464=15-17 | | 359b=10 | 359b=10 | | *461=13 | 461=13 | | | | | +- Carr - other16 | | 464=15-17 | | 359b=10? | | 461=13 | | *458=17 | | | +- Kilg./Rober. -+- Roberts - other24 | | 464=15-17 | 464=15-17 | | 359b=10 | 359b=10 | | *CDYb=37 | CDYb=37 | | *534=14 | 534=14 | | | *570=19 | | | *442=11 | | | | | +- Kilgour - other23 | | 464=15-17 | | 359b=10 | | 534=14 | | *CDYb=36 | | | +- Cerr - other11 | 359b=10 | *464=15-16-17 | +- McIntire - other12 | 464=15-17 | *448=19 | +- Kerr - other10 *464=15

Tree no. 2: This is the tree that results from assuming the unknown common ancestor had DYS359b=10 and DYS464=15-17. Since most of the Carr's have these values, this tree has the Carr's on the trunk and the Hige's as a branch.

unknown name ---+- Hoge/Carr ----+- Carr et.al. --+- Carr ---------+- Carr - other09/14 390=24 | | | 458=16 | | +- Carr - other15 359b=10 | | *439=11 448=20 | | 464=15-17 | +- Carr/Middl. --+- Middleton - other28 460=11 | | *461=13 | 461=13 GATAH4=11 | | | 570=18 | | +- Carr - other16 CDYa=36 | | 461=13 CDYb=38 | | *458=17 442=12 | | 534=15 | +- Kilg./Rober. -+- Roberts - other24 461=12 | | *CDYb=37 | CDYb=37 439=13 | | *534=14 | 534=14 | | | *570=19 | | | *442=11 | | | | | +- Kilgour - other23 | | 534=14 | | *CDYb=36 | | | +- Carr - other07 | | *CDYa=37 | | *534=16 | | | +- Cerr - other11 | *464=15-16-17 | +- Carr/Hoge/+ --+- McIntire - other12 *359b=9 | 359b=9 | *448=19 | +- Kerr - other10 | 359b=9 | *464=15 | +- Hoge/Carr ----+- Carr - other13 359b=9 | 359b=9 *464=15-17-18 | 464=15-17-18 | +- Hoge/Hogue ---+- Hoge/Hogue - NJ1682a/b/c, GA1820a | 359b=9 | 359b=9 | 464=15-17-18 | 464=15-17-18 | | | +- Hogue - NJ1682d | 359b=9 | 464=15-17-18 | *390=23 | +- Haml./McLean -+- McLean - other08 359b=9 | 359b=9 464=15-17-18 | 464=15-17-18 *CDYb=37 | CDYb=37 | *460=10 | +- Hamilton - other06 359b=9 464=15-17-18 CDYb=37 *GATAH4=12 *534=14

Counting the number of asterices in each of these two trees shows 20 mutations. We have tried other alternatives but have not found any other tree with less than 21 mutations. However, we cannot claim that our search has been exaustive and we cannot rule out the possibility that persistance will reveal other results. Suggestions and comments are welcome and encouraged.

We are not able to construct a tree with only 18 mutations. The difficulty is with CDYb and DYS534. These two markers show the same mutation in members that we had separated earlier based on the mutations in DYS359b and DYS464. We cannot escape the conclusion that CDYb and DYS534 mutated twice in independant branches. Note that CDYb has the highest mutation rate and DYS534 is higher than most of the others.

Some comments:

A tree of this sort is intended to show the sequence of mutations and gives an indication of which lines descend from a common ancestor more recent than the ancestor who was the common ancestor of all of them. It does not give an accurate representation of the actual family tree because we have no way of knowing whether any given motation was within the most recent generation of the DNA donor or was many generations ago.

It appeares that the two main surname branches, (1) Hoge/Hogue and (2) Carr/Kerr/Cerr/Karr, have equal right to claim to be the main branch but, Kilgour, Roberts, McIntire, and Middleton branch off from the Carr et.al. branch and Hamilton/McLean branck off of one or the other of the two possible main branches.