Haplogroup I2b KELLEY / KELLY Lineages and Results of Y-DNA Testing

Table of Contents

Diana, Goddess of the Hunt — for Ancestors!

Every-Name Index

Lineages and Results of Y-chromosome DNA Testing for Haplogroup I2b KELLEY / KELLY
(last updated January 2012)

INTRODUCTION

KELLY and KELLEY are common surnames, so it is no surprise that the KELLY DNA Project has many members. Trying to compare results in a table containing all the project's members can be a challenge. Therefore, as an aid to research on Haplogroup I2b KELL(E)Y, I've compiled a table of all the test subjects I could find who were Haplogroup I2b, not just in the project at FTDNA (FamilyTreeDNA), but any test data I could find elsewhere, as well. In tables below, I have grouped matching individuals and given as much lineage information as was supplied or that I could uncover.

I would like to encourage anyone to please not hesitate to contact me (address in the navigation bar at the bottom of the page) if you can add to the table, especially to improve the lineages or correct any errors I've made.

I am hosting a semi-private KELLEY-DNA-I2b mailing list. If you are a KELL(E)Y researcher interested in joining, please email me with your request to subscribe (see Contact link in navigation bar below).

Subclades of HAPLOGROUP I

Haplogroup I is divided into two groups: I1 and I2, with I1 being the most common of the two. I1 is concentrated in northwestern Europe, especially Scandinavia. I2a is concentrated in southeastern Europe, while the distribution of I2b parallels that of I1, without the concentration in Scandinavia (see distribution map, and see also the entry for Haplogroup I2 at Wikipedia). Immediately below is a table comparing the major versions of the subclades of Haplogroup I as defined by SNP mutations; these charts are undergoing revisions as new SNP mutations are discovered. I show only subclade I2b in detail because that is our interest here:

(An equal sign in the ISOGG column means the value is the same as FTDNA.)

Mutations Standard

FTDNA ISOGG

M170 P19 L41 M258 P212 P38 PS76 U179 I =

M253 I-1 =

P215=M438=S31 L68 I-2 =

L460 I-2a

P37.2 I-2a I-2a1

M26 L158 L159.1=S169.1 I-2a1 I-2a1a

M161 I-2a1a I-2a1a-1a

L160 I-2a1b I-2a1a-1

M423 L178 I-2a2 I-2a1b

P41.2=M359 I-2a2a I-2a1b-1

L161 I-2a2b I-2a1b-2

L233 I-2a3 I-2a1c

P217=S23 L181 L35=S150 L37=S153 P214=M436=S33 P216=S30 P218=S32 I-2b I-2a2

M223 L34=S151 L36=S152 L59 P219=S24 P220=S119 P221=S120 P222=S118=U250 P223=S117 | L368 L622 I-2b1 I-2a2a

M284 I-2b1a I-2a2a-1

L126=S165 L137=S166 I-2b1a-1 I-2a2a-1a

M379 I-2b1b I-2a2a-2

P78 I-2b1c I-2a2a-3

P95 I-2b1d I-2a2a-4

L38=S154 L39=S155 L40=S156 L65.1=S159.1 | L272.3 I-2b2 I-2a2b

L415 L416 L417 I-2b

L596 L597 I-2c

VARIETIES of HAPLOGROUP I

The dearth of SNP mutations defining subclades of Haplogroup I has led Nordtvedt to create varieties based on STR mutations (see his Excel spreadsheet [link died] for modals of these varieties). For a table showing his modal haplotypes for Haplogroup I varieties with the markers in FamilyTreeDNA order, please also see my HTML transcription of Nordtvedt's spreadsheet. To determine the varieties of our KELL(E)Y famlies, I depended on Jim Cullen's World Haplogroup & Haplo-I Subclade Predictor, which is based on Nordtvedt's research. Results are shown in the discussion below the data tables.
The reason you do not find modal values for four of the markers (viz., 567, 750, CDYa, CDYb) in the tables below is that Nordtvedt's work is based largely on the database at SMGF (the Sorensen Molecular Genealogy Foundataion), and SMGF does not test these markers. In all cases, I'm assuming the haplogroup subclade has been deduced from the haplotype. If any of you have been deep SNP tested, please let me know the result.

HAPLOGROUP I2b KELLEY/KELLY/etc. FAMILIES

So far, there are six distinct families in three varieties and two different subclades of Haplogroup I2b. Despite having different earliest known ancestors, everyone listed in the same group is closely related. Conversely, people in different groups are definitely not closely related. "Continental" here refers to the European continent; "Isles" refers to the British Isles.

Subclade Variety Group Earliest Known Ancestors _n_

I2b1
a.k.a.
I-M223 Continental-2a A William O'KELLY (c1600- ) emigrant from Ireland to VA in 1654 2

James O'KELLEY (c1660s-c1708) of Kent Co., MD 1

Richard KELLEY (c1750- ) emigrant from Ireland to the southeastern US 3

Joel KELLEY (c1775- ) of Fairfield Co., SC, and Bradley Co., TN 1

James M. KELLY (c1800- ) of Marion Co., TN 1

Edward KELLEY (c1799-) of SC 1

BLEVINS [also appears to have an NPE in his patrilineal line] 1

Continental-2b B William KELLEY (1715-1796) of Baltimore Co., MD 2

Moses KELLEY (1754-) of Baltimore Co., MD, and Somerset Co., PA 2

Thomas KELLEY (1778-1865) of MD and Huntingdon Co., PA 1

Unknown 1

I2b1a1
a.k.a.
I-L126 Isles-Scotland C John KELLY (1821- ) of Ireland 1

Edward Thomas KELLY (1874-1929) of Youngstown and Cleveland, OH 1

D Raymond KELLY (1887/8- ) of MN and Milwaukee, WI 1

E Michael KELLY (c1826- ) of co. Leitrim, Ireland, and Brooklyn, NY 1

F KELLY of Ireland 1

RECOMMENDATIONS

I administer six DNA projects based at FamilyTreeDNA — though not the KELLY project — and experience has taught me that 12 and 25 markers are simply not enough to determine with certainty whether individuals are closely related or not. I urge everyone to upgrade to at least 37 markers, preferably to 67. This upgrade is especially needed for those whose haplotype is modal or near modal at 12 or 25 markers because the more common your haplotype is, the more likely it is that you'll be getting matches that are merely coincidental and not actually indicative of relationship.

GENETIC DISTANCE

The Genetic Distance (GD) between two individuals is the number of mutation events that separate them. Most of the time, GD can be easily determined by a simple count of their numerical difference. There are certain mutations events, however, that can change the count by more than a single digit in one mutation event. Where these events have occurred among the test subjects, I explain them in the discussion below the tables. The genetic distances in the tables are between the individual test subject and their own family's modal haplotype.

ASSESSING RELATEDNESS

In the tables below, I have included the modal haplotypes for their I2b varieties. I have determined these varieties using Jim Cullen's Haplogroup I Subclade Predictor, and the results of the prediction are given in the discussion below the tables. The important markers to note in the tables are not the ones where they are modal, because there's little significance to being modal, but where individuals are different from the modal. In cases where we have more than one individual in the family tested, the most important markers are the ones where they are not only not modal, but are consistent in their differences from the modal. These markers are the "signature" of the family. One important reason to test 67 markers is to determine all the signature markers for the family, especially when the family has few signature markers in the lower panels. (I discuss how to interpret Y-DNA STR test results further on this web page.)

MY MOTIVATION

I descend from two KELLEY women, both of Baltimore Co., MD: the Sarah KELLEY who married Elijah CORBIN, and the Margaret KELLEY who married their son, Joshua CORBIN. Sarah is alleged to be a daughter of William KELLEY (1715-1796) and his second wife, Eleanor CORBIN, while Margaret is alleged to be a granddaughter of James KELLEY (1710-<1780) & Prudence LOGDSON. James and William are alleged sons of James KELLEY (1680-1750) of co. Cork, Ireland, and Baltimore Co., MD. None of these connections has yet been proven on paper, but these KELLEYs and CORBINs migrated together from Baltimore [now Carroll] Co., MD, to Huntingdon Co., PA, so I've no doubt my two KELLEY females connect to William and/or James, one way or another. These are the Group B KELLEYs below.
Update (1 May 2014): I and four of my cousins have Family Finder matches with three descendents of William & Honora (O'FLYNN) LOGSDON, parents of Prudence LOGSDON. There seems no other likely mutual connection than this one in our pedigrees, so perhaps it's time to accept that Margaret (KELLEY) CORBIN is the daughter of James & Prudence (LOGSDON) KELLEY.

SOURCES, RESOURCES, and ACKNOWLEDGMENT

1. The KELLEY/KELLY DNA Project and its sister site at FTDNA (as of 4 Jan 2012).
2. Jim Cullen's World Haplogroup and Haplogroup-I Subclade Predictor.
3. Clusters from the Haplogroup I2b1/M223 Project.
4. The Ysearch database (as of 20 Jan 2009).
5. The SMGF (Sorenson Molecular Genealogy Foundation) database (as of 7 Jul 2009).
Some individuals occur in both the FTDNA and Ysearch databases, so will have both an FTDNA Kit# and a Ysearch UserID. I believe I've managed to tell which ones are duplicated, so that I have the correct Kit# associated with the correct Ysearch UserID, but if you think I've erred, please correct me. The individuals with "A-DNA" or "A-#####" in the Kit# column were tested at Ancestry-DNA and found at Ysearch; the "SGMF" individual was found in the Sorenson Molecular Genealogy Foundation database.
I would like here to acknowledge the cooperation and support of Raymond KELLY, administrator of the KELLEY Y-DNA Surname Project.

To view more of the page without scrolling, temporarily reduce the text size or page size in your browser.
Red labels indicate markers that typically mutate more frequently than those labeled in black. Markers 464 and 570 mutate most rapidly of all.

Haplogroup I2b1 = I-M223

Continental-2a Cluster
Group A (= Group 40 at the Kelley Project) - KELLEY of Ireland and the Southeastern United States
To view lineages, please scroll to right.

GD
(cumulative)

Surname

FTDNA
Kit #

Ysearch
UserID

Haplotype — as determined by STR testing

Lineage

Markers 1-12

Markers 13-25

Markers 26-37

Markers 38-67

Advanced Markers

at
12

at
25

at
37

at
67

3
9
3

3
9
0

19
/
3
9
4

3
9
1

a
|
3
8
5

b
|
3
8
5

4
2
6

3
8
8

4
3
9

i
|
3
8
9

3
9
2

ii
|
3
8
9

4
5
8

a
|
4
5
9

b
|
4
5
9

4
5
5

4
5
4

4
4
7

4
3
7

4
4
8

4
4
9

a
|
4
6
4

b
|
4
6
4

c
|
4
6
4

d
|
4
6
4

e
|
4
6
4

f
|
4
6
4

4
6
0

H4
|
G
A
T
A

IIa
|
Y
C
A

IIb
|
Y
C
A

4
5
6

6
0
7

5
7
6

5
7
0

a
|
C
D
Y

b
|
C
D
Y

4
4
2

4
3
8

5
3
1

5
7
8

a
|
S1
3
9
5

b
|
S1
3
9
5

5
9
0

5
3
7

6
4
1

4
7
2

S1
4
0
6

5
1
1

4
2
5

a
|
4
1
3

b
|
4
1
3

5
5
7

5
9
4

4
3
6

4
9
0

5
3
4

4
5
0

4
4
4

4
8
1

5
2
0

4
4
6

6
1
7

5
6
8

4
8
7

5
7
2

6
4
0

4
9
2

5
6
5

4
6
1

4
6
2

A10
|
G
A
T
A

C4
|
6
3
5

B
0
7

4
4
1

4
4
5

4
5
2

4
6
3

I-M223-Cont2a Modal Values

4H6C9

Modal values per Nordtvedt

Family Modal Values

5KS3B

Modal values (n=12 at 12; n=11 at 25; n=9 at 37; n=6 at 67)

KELLY

133887

Pvt⁹, William Gillespie⁸, Henry James⁷, William Howell⁶, Vincent⁵, Richard⁴, Benjamin³, Benjamin², James¹ O'KELLY (c1660s-c1708) of Kent Co., MD

KELLEY

142553

Pvt⁶, William Herschel⁵, Benjamin Franklin⁴,

Leander Homer³,

Benjamin Franklin², Richard¹ KELLEY (c1750- ), emigrant from IRL to southeastern US

KELLEY

148276

48CUG

Pvt⁸, Garry B.⁷, Boyd Ross⁶, John Dearton⁵, John Smith⁴,

Richard Yancey³,

KELLEY

SMGF

Pvt⁷, Wendell Orville⁶, Richard Cassius Marion⁵, William Henry⁴,

KELLEY

31687

VVH3Q

Pvt⁵, Jesse James⁴, Hiram Douglas³, Elijah², Joel¹ KELLEY (c1775- ) of Fairfield Co., SC, and Bradley Co., TN

KELLEY

157867

Pvt…

BLEVINS

171547

Pvt…

KELLEY

60671

YPAZ8

Pvt⁶, Ernest Richey⁵, Elbert Vernon⁴, James Riley³, Riley A.², James M.¹ KELLY (c1800- ) of Marion Co., TN

KELLY

148174

4Q5R8

Pvt¹², Richard Alexander¹¹, Alexander Doniphan¹⁰, Granville James⁹, John Payne⁸, John⁷, James⁶, Alexander⁵,

Capt. John⁴, Matthew³, John² KELLY, William¹ O'KELLY (c1600?- ), emigrant from IRL to VA in 1654

KELLY

A10172555

YU84S

Pvt¹³, Pvt¹², John Thomas¹¹, John Thomas¹⁰, Sidney Mann⁹, John⁸, Reuben⁷, John⁶, Matthew⁵,

KELLEY

N29340

Pvt⁴, William Levi³, William Thomas², Edward¹ KELLEY (c1799-) of SC

Using Jim Curren's Subclade Predictor, the modal haplotype for Group A gives these probabilities:
I-M223-Cont2a =>44%, I-M223-Cont2b =>35%, and I-M223-Cont1a =>18%.

#148276 has been deep SNP tested, with these results:

P217+

M223+

M284- M379- P78- P95-

The positive result for P217 proves he is I2b; the positive result for M223 proves he is I21b. The negative results prove he is not a subclade of I21b, but rather root/ancestral I21b*.

Ten of the eleven members of Group A match each other 12/12, but we can't be certain the match for one of them will hold up until he upgrades to at least 37 markers because his 12-marker haplotype differs from the modal haplotype for Cont2a by only one mutation. In other words, it's a common haplotype.

Based on just 12 markers, it would not be thought that #N29340 belonged in this group, at all, especially as a 10/12 match tends to get worse, not better, as more markers are tested. In this case, however, his match improved at more markers, so it's another lesson in the importance of testing to 37 markers and the serious inadequacy of testing just 12.

DYS464 is a multi-copy marker where the order of the alleles is not known, so, by convention, they are listed lo-hi. In the case of the Group A individuals, the modal values shown on the member pages are in this lo-hi order: 11,11,14,14. However, so as not to over-estimate the genetic distance when comparing haplotypes, the alleles of DYS464 should be re-ordered so as to produce the closest match. In this case, because the difference is in the value of 15 for the modal and 11 for our family, I've reordered our subjects' alleles to 11,14,14,11, so that they differ on only one allele, not two. And even though the count between 15 and 11 differs by four, it's considered a single mutation event because what has happened is that the 11 allele over-wrote the 15 allele in a what's known as a RecLOH event (a Recombinant Loss Of Heterozygosity — a loss of genetic diversity during cell division).

#148174 has two additional alleles at DYS464, a mutation that is probably fairly recent as no one else, so far, shares it. Having more than four alleles is a rare condition occurring in only about 1.5% of people tested. As expected, his results are reported lo-hi on his member page: 11,11,11,11,14,14. Again, I have re-ordered his alleles to produce the least difference from other family members: 11,14,14,11,11,11. Ysearch scores this difference as a genetic distance of two, but it's entirely possibly, in fact probable, that the two additional copies appeared in one mutation event by a duplication of both 11 alleles at once. The testing of cousins could prove whether this condition is the result of one mutation event or two, but it is most likely one, so I have scored it as a GD of 1, not 2.

Because Marker CDY is prone to multi-step mutations, it is probable that the changes from 43 to 41 and from 43 to 40 at CDYb are each single mutation events, so I have tallied these differences as single mutation events, not two or three events. Again, the testing of cousins can tell us how many events really took place.

It is probably significant that #60671 and #148174 share the mutation at CDYab from 34,43 to 35,41 (bright green table cells). I say probably because CDYab is a volatile marker, and mutltiple mutations in short time frames are possible. Even so, what it probably means is that the two share a near common ancestor, downstream of their progenitor, but upstream of #148174's mutation at DYS464 and upstream of #60671's mutations at DYS617 and DYS572. Testing of more cousins can determine the locations of these mutations.

In translating Ancestry-DNA values to FTDNA values, these conversion factors are used: DYS 441, subtract 1; DYS 442, subtract 5; GATA-H4, subtract 11; Y-GATA-A10, subtract 2. SMGF values do not need conversion, provided they are extracted from the SMGF database using the FTDNA lab standard.

Continental-2b Cluster
Group B (= Group 39 at the KELLEY Project) — KELLEY of Ireland, then Maryland and Points Westward
To view lineages, please scroll to right.

GD
(cumulative)

Surname

FTDNA
Kit #

Ysearch
UserID

Haplotype — as determined by STR testing

Lineage

Markers 1-12

Markers 13-25

Markers 26-37

Markers 38-67

Advanced Markers

3
9
3

3
9
0

19
/
3
9
4

3
9
1

a
|
3
8
5

b
|
3
8
5

4
2
6

3
8
8

4
3
9

i
|
3
8
9

3
9
2

ii
|
3
8
9

4
5
8

a
|
4
5
9

b
|
4
5
9

4
5
5

4
5
4

4
4
7

4
3
7

4
4
8

4
4
9

a
|
4
6
4

b
|
4
6
4

c
|
4
6
4

d
|
4
6
4

4
6
0

H4
|
G
A
T
A

IIa
|
Y
C
A

IIb
|
Y
C
A

4
5
6

6
0
7

5
7
6

5
7
0

a
|
C
D
Y

b
|
C
D
Y

4
4
2

4
3
8

5
3
1

5
7
8

a
|
S1
3
9
5

b
|
S1
3
9
5

5
9
0

5
3
7

6
4
1

4
7
2

S1
4
0
6

5
1
1

4
2
5

a
|
4
1
3

b
|
4
1
3

5
5
7

5
9
4

4
3
6

4
9
0

5
3
4

4
5
0

4
4
4

4
8
1

5
2
0

4
4
6

6
1
7

5
6
8

4
8
7

5
7
2

6
4
0

4
9
2

5
6
5

4
6
1

4
6
2

A10
|
G
A
T
A

C4
|
6
3
5

B
0
7

4
4
1

4
4
5

4
5
2

4
6
3

I-M223-Cont2b Modal Values

8/9

Group B Family Modal Values

SUBEF

n=6 at 12; n=4 at 25; n=3 at 37

KELLEY

N33799

Pvt⁸, Gilbert Andrew⁷, Everett Perry⁶, John George⁵, Levi⁴, John³, William²,

William¹KELLEY (1715-1796) of Baltimore Co., MD

KELLEY

N2277

Pvt⁸, Jesse Arthur⁷, Henry Clay⁶, George Robert⁵, James McCarty⁴, Gideon³, Joshua²or_Zachariah²,

KELLEY

153312

Q6VT9

Pvt⁶, Edward Jackson⁵, Edward Jackson⁴, David D.³, Andrew Jackson²,

Thomas¹ KELLEY (1778-1865) of MD and Huntingdon Co., PA

KELLEY

107540

Pvt⁶, Marion Stanley⁵, George Washington⁴, Joseph Edward³, William Allen²,

Moses¹, KELLEY (1754-) of Baltimore Co., MD, and Somerset Co., PA

KELLEY

101820

77AEQ

Pvt⁶, Alva⁵, John Morgan⁴, Lee³, Elias²,

KELLEY

A-DNA

JDZEC

Pvt…

Using Jim Curren's Subclade Predictor, the modal haplotype for Group B gives these probabilities:
I-M223-Cont2b =>58%, I-M223-Cont1 =>38%, and I-M223-Cont2c =>3%.

#N2277 has been deep SNP tested, with the result that his most downstream positive SNP is M223, confirming that he is Haplgroup I21b.

The "signature markers" for this family are highlighted in green. Values of 9,10 at DYS459a,b are actually the most common values, overall, but a value of 9 at DYS459a happens to be uncommon in Haplogroup I, where the values are usually 8,9. A value of 13 at DYS456 occurs in only 1.8% of people tested (n=27,000+). "Private markers" are highlighted in purple.

The members of Group B who have tested over 25 markers are definitely related, with #N33799 being the modal "in-betweener" who unites the others. Two of these individuals have a paper descent from William KELLEY (1715-1796) of Baltimore Co., MD, a major progenitor of KELLEY in the U.S.

I have created an account at Ysearch for this family's modal haplotype (SUBEF). The only tightly matching KELL(E)Ys are the ones above who have uploaded to Ysearch, but the family's closeness to the modal haplotype of Cont2b means they have many near matches in other surnames. Testing 67 markers should pull them away from the pack. They have no significant matches in the SMGF database. (Searches as of 11 Jul 2009.)

Many claim William (1715-1796) is son of James (c1680- ) of co. Cork, Ireland, but I have yet to find any evidence connecting them.

One member claims Moses (1754- ) is a son of William (1715-1796), but I have yet to find any evidence that William had a son, Moses.

The member descended from Gideon (c1780-1830s) claims Gideon is s/o Zachariah (1760-1782), s/o William (1715-1796), but I have yet to find any evidence connecting Gideon to Zachariah or, in fact, any evidence that Zachariah survived to marry (note his age at death). There is evidence that Gideon is at least related to Joshua KELLEY, older brother of Zachariah, because they are listed next to each other in the 1810 census. It is possible that Joshua is Gideon's uncle, but on the face of it, he appears to be Gideon's father. Whichever, I don't think there's any doubt he's William's grandson.

Haplogroup I2b1a1 = I-L126

IslesSc Cluster (ungrouped at the Kelley Project)
To view lineages, please scroll to right.

G
R
O
U
P

GD
(cumulative)

Surname

Kit#

Ysearch
UserID

Panel 1 — Standard Markers 1-12

Panel 2 — Standard Markers 13-25

Panel 3 — Standard Markers 26-37

Panel 4 — Standard Markers 38-67

Panel 5 — Standard Markers 68-111

Known Lineage

STR

PP3

PP5

MP6

MP7

MP8

MP9 [mixed]

MP10 [EA]

MP11 [SMGF]

MP12 [all new]

MP13 [mixed]

at
12

at
25

at
37

at
67

at
111

3
9
3

3
9
0

19
/
3
9
4

3
9
1

a
|
3
8
5

b
|
3
8
5

4
2
6

3
8
8

4
3
9

i
|
3
8
9

3
9
2

ii
|
3
8
9

4
5
8

a
|
4
5
9

b
|
4
5
9

4
5
5

4
5
4

4
4
7

4
3
7

4
4
8

4
4
9

a
|
4
6
4

b
|
4
6
4

c
|
4
6
4

d
|
4
6
4

4
6
0

H4
|
G
A
T
A

IIa
|
Y
C
A

IIb
|
Y
C
A

4
5
6

6
0
7

5
7
6

5
7
0

a
|
C
D
Y

b
|
C
D
Y

4
4
2

4
3
8

5
3
1

5
7
8

a
|
S1
3
9
5

b
|
S1
3
9
5

5
9
0

5
3
7

6
4
1

4
7
2

S1
4
0
6

5
1
1

4
2
5

a
|
4
1
3

b
|
4
1
3

5
5
7

5
9
4

4
3
6

4
9
0

5
3
4

4
5
0

4
4
4

4
8
1

5
2
0

4
4
6

6
1
7

5
6
8

4
8
7

5
7
2

6
4
0

4
9
2

5
6
5

7
1
0

4
8
5

6
3
2

4
9
5

5
4
0

7
1
4

7
1
6

7
1
7

5
0
5

5
5
6

5
4
9

5
8
9

5
2
2

4
9
4

5
3
3

6
3
6

5
7
5

6
3
8

4
6
2

4
5
2

4
4
5

A10
|
G
A
T
A

4
6
3

4
4
1

1B
07
|
GG
AA
T

5
2
5

7
1
2

5
9
3

6
5
0

5
3
2

7
1
5

5
0
4

5
1
3

5
6
1

5
5
2

7
2
6

6
3
5

5
8
7

6
4
3

4
9
7

5
1
0

4
3
4

4
6
1

4
3
5

IslesSc Cluster Modals

KELLY

N48960

Pvt⁵, John F.⁴, John F.³, Charles², John¹ KELLY (1821- ) of Ireland

KELLY

DNA-A

NZ9CE

Pvt³, Pvt², Edward Thomas¹ KELLY (1874-1929) — of Youngstown and Cleveland, OH

KELLY

N15779

Pvt², Raymond¹ KELLY (1887/8- ) of MN and Milwaukee, WI

KELLY

N59513

Pvt⁴, Paul Brendan³, Francis², Michael¹ KELLY (c1826- ) of Co. Leitrim, Ireland, and Brooklyn, NY

KELLY

152126

Pvt… KELLY — of Ireland

Using Jim Cullen's Subclade Predictor, the haplotypes for these groups yield these probabilities:
Groups C and F: I-M284-Isles/Sc =>100%
Group D: I-M284-Isles/Sc =>96% I-M223-Cont2b =>3%
Group E: I-M284-Isles/Sc =>68% I-P78-Cont3a =>24% I-M223-Cont1a =>7%

The probabilities are lower for Groups D and E because they've only tested 12 markers.

FTDNA shows these individuals as Haplogroup I2b1. Cullen's haplogroup subclade predictor indicates their most downstream SNP would be expected to be M284, which would make them I2b1a. Nordtvedt's "IslesSc" variety has a high concentration in the British Isles, expecially Scotland.

The genetic distances are within group only; so, as yet, they are only meaningful for Group C. The differences between groups are enough to indicate that the four groups are probably not related in genealogical time, but that determination would be much more clear if everyone would upgrade to 67 markers.

#N48960 and the individual tested at Ancestry-DNA have a low-level match at 25/27. Because #N48960 is very close to the modal value for IslesS, I would encourage the individual tested at A-DNA to join the KELLY project at FTDNA, so he can compare a full 37 markers. I've seen a 23/25 match drop to 28/37, so we cannot be certain, at this point, that the two are really closely related. I would, in fact, encourage both to upgrade to 67 markers.

#N15779 has fewer than a dozen full or near matches at Ysearch and none in KELL(E)Y. As with the above individual, his testing needs to be upgraded to make certain he is not a high level match with one of these other surnames, to rule out having an NPE. However, it has come to my attention that the test subject is deceased; and, certainly, his email address is. If you are reading this and became heir to his property, please contact FTDNA to get yourself assigned as owner and contact person for his account.

#N59513 has dozens of full and near matches at Ysearch, but none in KELL(E)Y. Although it would seem that if he has no near matches with KELL(E)Y that there would be no urgency to upgrade to more markers, he does need to upgrade to make certain he is not a high level match with one of these other surnames — to rule out that he has an NPE in his line.

#152126

What constitutes a match?
Matches in other surnames are usually mere coincidence, so please ignore them — I'll let you know when you shouldn't!

For 12 markers:	9 or less is a non-relative;	for 10-12 markers,	please see this table compiled by FTDNA.
For 25 markers:	21 or less is a non-relative;	for 22-25 markers,
For 37 markers:	31 or less is a non-relative;	for 32-37 markers,
For 67 markers:	59 or less is a non-relative;	for 60-67 markers,
For 111 markers:	100 or less is a non-relative;	for 101-111 markers,
For any test:	0 matching markers,		please contact NASA.