This is an email from GenealogiaMolecular.com
Message: The Science of Molecular Genealogy
By Ugo A. Perego; Ann Turner, M.D.;
Jayne E. Ekins; and Scott R. Woodward, Ph.D.
Molecular science can help genealogists uncover previously unknown family relationships,
verify or refute claims to ancestry, and shed light on questions that have puzzled
family historians for years.
All individuals carry a record of their ancestors in a complex chemical
compound found inside almost every cell. Analysis of this molecule—
deoxyribonucleic acid (DNA)—can help genealogists trace male- and
female-line ancestors, prove and disprove relationships, reveal undocumented
illegitimacies and adoptions, and identify familial ethnic and geographic origins.
DNA is packaged in threadlike structures called "chromosomes." Humans
receive twenty-three chromosomes from each parent and, in turn, give half of
their own DNA to each of their children. Parents, therefore, funnel a molecular
record of their ancestors to their descendants.
More than 99 percent of each person's DNA is identical to that of all other
people. This shared inheritance defines humans, yet the remaining 1 percent
contains enough variation to make each person unique. The DNA of two closely
related people has more similarities than that of distant cousins. Consequently,
similarities and differences in DNA can show how closely individuals are related.
Molecular genealogists—also called "genetic genealogists"—test DNA
samples from living individuals. Used in isolation, DNA test results have little
value for family historians. Combined with documentary genealogical research,
however, DNA evidence can help researchers identify ancestors and reconstruct
family histories and lineages. Suppose, for example, that research reveals a candidate
for a male ancestor's father but does not prove the relationship. If DNA
samples from living male-line descendants of both men are different, they will
disprove the hypothesis. If the samples match, however, the DNA alone does not
© Ugo A. Perego, ugo@smgf.org; Ann Turner, M.D., DNACousins@aol.com; Jayne E. Ekins, jayne@smgf
.org; and Scott R. Woodward, Ph.D., scott@smgf.org. Mr. Perego is director of operations for the Sorenson
Molecular Genealogy Foundation (SMGF), where he supervises the collecting of genetic and genealogical data
for the foundation's worldwide database. In the past five years, he has written a dozen articles and given more
than one hundred lectures on molecular genealogy. Dr. Turner founded the GENEALOGY-DNA mailing list
at RootsWeb and co-authored (with Megan Smolenyak) Trace Your Roots with DNA: Using Genetic Tests to
Explore Your Family Tree (Emmaus, Pa.: Rodale, 2004). Ms. Ekins, a research scientist with SMGF, coordinates
genetic data production and analysis and performs original research. Dr. Woodward, whose work has been
featured internationally, is chief scientific officer for SMGF and principal investigator for the molecular
genealogy research project. His research interests include reconstructing ancient and modern genealogies using
DNA techniques on samples worldwide, tracing human population movements by following gene migrations,
and analyzing DNA found in ancient manuscripts.
NATIONAL GENEALOGICAL SOCIETY QUARTERLY 93 (DECEMBER 2005): 245–59
prove a father-son relationship but, in combination with documentary evidence,
it could make the case persuasive.1
GENEALOGICALLY USEFUL PATHWAYS OF GENETIC INHERITANCE
Each parent contributes approximately half of his or her child's DNA. Scientists
usually cannot identify which parent provided which part of the child's
DNA without testing one or both parents. The parts that researchers can identify,
however, allow them to answer many genealogical questions: the Y chromosome,
which is found in each cell's nucleus in males only, and mitochondrial DNA
(mtDNA), which is found in each cell's cytoplasm. See figure 1.
The Paternal Lineage Pathway
Sons receive a Y chromosome, usually unchanged, from their fathers. Occasionally,
however, a slight alteration (called a "mutation") will occur in a random
male's Y chromosome. A man with such an altered Y chromosome will pass it to
his sons and they to their sons. Subsequently, all of their male-line descendants
will pass that slightly altered Y chromosome to their sons. Further random mutations
may occur occasionally in later generations. Thus, every living male's Y
chromosome today carries a cumulative history of many small changes that have
occurred in his paternal lineage over hundreds of generations and thousands of
1. Several recent books cover the basic biology of genetics and DNA as they apply to genealogical testing.
See, for example, Chris Pomery, DNA and Family History (Toronto, Ont.: Dundurn, 2004); Thomas H.
Shawker, Unlocking Your Genetic History (Nashville, Tenn.: Rutledge Hill, 2004); and Smolenyak and Turner,
Trace Your Roots with DNA.
National Genealogical Society Quarterly 246
years. Because different changes occurred in different males over the millennia,
their male descendants bear Y chromosomes with distinctive patterns, called
"haplotypes," which can differentiate their families and ancestors.
The Y chromosome is useful for answering genealogical questions because it
passes intact from generation to generation and its inheritance follows surnames
in many western and some nonwestern societies. All male-line descendants of the
same male ancestor—typically those with the same surname in these societies—
will have the same or a very similar Y chromosome.2 For example, residents of
Tristan da Cunha, which has genealogical records dating from 1816, have eight
Y-chromosome haplotypes corresponding to those of seven of the island's
founders, whose surnames the residents bear, and an apparent visitor with an
unknown surname.3 In cases where the Y-chromosome haplotype did not correspond
to the surname, it indicated ancestry more accurately than documentary
genealogy or oral history.
Searching a database of Y-chromosome haplotypes paired with surnames can
enable genealogists to identify relatives and disprove erroneous lineages. Males
with the same surname and different haplotypes probably descend from different
lines bearing the same surname. Conversely, similar haplotypes of males with
different surnames might indicate adoption, illegitimacy, or other situations
where names may have been altered somewhere in a male-line descent. For
example, men with Lorentz and Lawrence surnames and the same Y-chromosome
haplotype very likely descend from the same male-line ancestor.4
Two studies popularized applying Y-chromosome analysis to genealogical research,
the highly publicized Jefferson-Hemings case and a study involving the
Jewish priestly class of Cohen:
• The question of whether Thomas Jefferson fathered some or all of his slave Sally
Hemings's children arose during his lifetime, and it is still the subject of debate
today. Jefferson left no male issue through his wife, but living male-line descendants
of his father's brother, who carried the same Y chromosome as Jefferson, were tested
to determine the Jefferson haplotype. Also tested were male-line descendants of
(1) Jefferson's brother-in-law John Carr (because of rumors that members of his
family had fathered Hemings's children), (2) Sally Hemings's son Thomas Woodson,
and (3) another Hemings son, Eston. Of the three lines, only the male-line
descendants of Eston Hemings carry the Jefferson Y-chromosome haplotype.5 The
2. Bryan Sykes and Catherine Irven, "Surnames and the Y Chromosome," American Journal of Human
Genetics 66 (March 2000): 1417–19; and Mark A. Jobling, "In the Name of the Father: Surnames and Genetics,"
TRENDS in Genetics 17 (June 2001): 353–57.
3. Himla Soodyall and others, "Genealogy and Genes: Tracing the Founding Fathers of Tristan da
Cunha," European Journal of Human Genetics 11 (September 2003): 705–9.
4. Ann Turner, "One or Many? Ann Turner Looks at the Role of DNA in the Study of Surname Origins,"
Family Chronicle 9 (March/April 2005): 46–49.
5. Eugene A. Foster and others, "Jefferson Fathered Slave's Last Child," Nature 396 (5 November 1998):
27–28.
The Science of Molecular Genealogy 247
DNA evidence alone does not prove that Jefferson was Eston's father, but it
complements evidence drawn from other sources.6
• Researchers found that a noticeable fraction of Jewish priests share a common
Y-chromosome haplotype, whether or not they are part of the far-flung
Ashkenazi or Sephardic communities.7 A later study found the same haplotype in
the Lemba of southern Africa, a tribe with customs reminiscent of Jewish practices
and an oral tradition that their ancestors came from the north by boat.8 Finding the
same haplotype in geographically dispersed groups implies descent from a single
common ancestor.
Businessman Bennett Greenspan hoped that the approach used in the Jefferson
and Cohen research would help family historians. After reaching a brick wall
on his mother's surname, Nitz, he discovered an Argentine researching the same
surname. Greenspan enlisted the help of a male Nitz cousin. A scientist involved
in the original Cohen investigation tested the Argentine's and Greenspan's cousin's
Y chromosomes. Their haplotypes matched perfectly. Furthermore, the haplotype
did not match any of two dozen samples collected by Greenspan to serve
as controls. Fortified by this demonstration that DNA could reflect a common
lineage, Greenspan founded a private company offering DNA tests for genealogical
purposes. His business was shortly followed by a half-dozen similar companies
in the United States and Europe.9
More than two thousand Y-chromosome surname studies are underway, some
with hundreds of participants.10 Family historians interested in joining a project
can find lists of active investigations on commercial testing company Web sites
and Ancestry.com and Genforum.com message boards.11 Many surname-project
Web sites report genetic findings, and genealogical periodicals are beginning to
carry case studies that include Y-chromosome analyses. Several examples demonstrate
different genealogical uses of Y-chromosome data:
• Hundreds of men named Wells participated in Y-chromosome testing. Genealogical
data collected prior to the project suggested twenty-four distinct families. The
6. Helen Leary, "Sally Hemings's Children: A Genealogical Analysis of the Evidence," and Thomas W.
Jones, "The 'Scholars Commission' Report on the Jefferson-Hemings Matter: An Evaluation by Genealogical
Standards," NGS Quarterly 89 (September 2001): 165–207 and 208–18.
7. Mark G. Thomas and others, "Origins of Old Testament Priests," Nature 394 (9 July 1998): 138–40.
8. Mark G. Thomas and others, "Y-chromosomes Traveling South: The Cohen Modal Haplotype and the
Origins of the Lemba—the 'Black Jews of Southern Africa'," American Journal of Human Genetics 66 (February
2000): 674–86.
9. Bennett Greenspan, "An Insider's Look at the Genealogy DNA Field," New England Ancestors
(Summer 2004): 21–23.
10. Bill Davenport, "Surname Projects: 'Over Fifty List'," World Families Network (http://worldfamilies.net/
over50list.html).
11. The largest companies with family DNA projects are DNA Heritage (http://www.dnaheritage.com),
Family Tree DNA (http://www.familytreedna.com), and Relative Genetics (http://www.relativegenetics.com).
A more complete listing can be found at Megan Smolenyak Smolenyak, "Genetealogy Resources," Genetealogy.
com (http://genetealogy.com).
National Genealogical Society Quarterly 248
Y-chromosome surname study, however, demonstrates that five presumed connections,
based on similar names, dates, and places, are separate lines.12
• Y-chromosome samples from just two people solved a problem that had baffled
researchers for years. Justin Howery and Fred Hauri, who believed that everyone
with a variant of their surnames descends from a man who lived in the 1400s in the
Swiss village of Beromuenster, could not document a family connection. Genetic
testing, however, revealed that both men have the same Y-chromosome haplotype,
even though their ancestors came to the United States from different countries in
different centuries.13
• Although all Smolenyaks seem to trace their ancestry to one small village in
Slovakia, they have four Y-chromosome haplotypes, indicating four distinct ancestral
lines.14
• The ancestral haplotype of Edmund Rice, who immigrated to Massachusetts in
1638, was established by matching DNA results from descendants of five different
sons. The testing also revealed a "non-paternity event"—possibly an unrecorded
adoption or illegitimacy—in one line of male descent.15
The Maternal Lineage Pathway
In addition to the DNA in the nucleus of most cells, DNA is also found in
structures called mitochondria in each cell's cytoplasm. See figure 1. Cells have
hundreds of mitochondria, each containing many DNA molecules called "mitochondrial
DNA" or "mtDNA." The mother's—but not the father's—
mitochondria are present in the fertilized egg that is the first cell of a new human
being. Thus, a mother passes her mtDNA to her sons and daughters. Her daughters,
but not her sons, pass their mtDNA to the next generation. Consequently,
mtDNA, inherited exclusively from mothers, passes intact from generation to
generation.
Just as with the Y chromosome, slight random changes in mtDNA molecules
over many generations result in different patterns or haplotypes. However, these
mutations occur less frequently than in Y chromosomes. The mutation rate has
been measured in Iceland, which has genealogical records covering many generations.
Only three mutations occurred in 705 "transmission events" (opportunities
for a mutation to occur between generations). Some of the residents with
matching mtDNA haplotypes were twelve generations removed from their common
female-line ancestor, who was born in 1560.16
12. Ken Wells, "Relative Advance: DNA Testing Helps Find Family Roots," Wall Street Journal, 6 March,
2003, page A1.
13. Justin Howery, "Howery DNA Project," message board posting, 17 November 2000, GENEALOGYDNA-
L Archives (http://archiver.rootsweb.com/th/read/GENEALOGY-DNA/2000–11/0974503831).
14. Megan Smolenyak Smolenyak, "DNA Testing Dispels a Genealogical Myth," Everton's Family History
Magazine 56 (May/June 2002): 44–48.
15. Robert V. Rice and John F. Chandler, "DNA Analyses of Y-chromosomes Show Only One of Three
Sons of Gershom Rice to be a Descendant of Edmund Rice," New England Ancestors 3 (Fall 2002): 50–51.
16. Sigrun Sigurgardottir and others, "The Mutation Rate in the Human mtDNA Control Region,"
American Journal of Human Genetics 66 (May 2000): 1599–1609.
The Science of Molecular Genealogy 249
The popularity of mtDNA for genealogical purposes followed the use of
mtDNA to confirm the identity of remains thought to be those of the wife and
children of Nicholas II, czar of Russia. The mtDNA extracted from the remains
matched that of living relatives who shared a common maternal line with the
czar's wife.17 Such testing can disprove relationships as well: the mtDNA of
Anna Anderson Manahan, who claimed to be Nicholas's daughter Anastasia, did
not match that of the czar's family.18
Ethnic and Geographic Pathways
"Autosomes" are twenty-two pairs of chromosomes that children inherit from
both parents, and the DNA they contain is called "autosomal DNA." Autosomes
do not include the sex chromosomes (X and Y) and mtDNA. In contrast to
mtDNA and Y-chromosome molecules, which parents pass intact to their children,
autosomal DNA comprises a random combination of both parents' genetic
makeup. Because each parent's autosomal DNA recombines, the half that each
transmits to a child is a mixture of the DNA that the contributing parents
received from their parents. Consequently, if neither parent's autosomal DNA is
studied, it is not possible to determine which parent or ancestor contributed any
segment of the child's autosomal DNA. In addition, the proportion of ancestral
genetic contribution to a descendant decreases with the number of generations
between the ancestor and descendant. Both the recombination of autosomal
DNA and its decreasing proportions over generations create challenges for using
autosomal DNA for genealogical purposes.
Autosomal DNA is the focus of techniques to determine origins more generally
than paternal and maternal lines allow. Genealogists can submit DNA
samples to a company that tests autosomal DNA to identify continental or
subcontinental origins.19 In a broad sense, individuals can use the results to
reconnect with family roots that may have been previously unknown to them.
Other research has focused on inferring ancestry more specific than broad
geographic or ethnic classifications. A recently proposed approach allows a participant's
assignment to a hierarchical set of populations. For example, the firstlevel
test results may imply European origins. Subsequent levels may narrow the
inference successively to the British Isles, a region in southwest Wales, and
perhaps an extended family from the area. Such inference of ancestry from all
17. Peter Gill and others, "Identification of the Remains of the Romanov Family by DNA Analysis,"
Nature Genetics 6 (February 1994): 130–35. Recent publications have challenged the findings on the basis of
difficulties involved in recovering and analyzing ancient DNA. See Alex Knight and others, "Molecular,
Forensic and Haplotypic Inconsistencies Regarding the Identity of the Ekaterinburg Remains," Annals of Human
Biology 31 (March–April 2004): 129–38.
18. Peter Gill and others, "Establishing the Identity of Anna Anderson Manahan," Nature Genetics 9
(January 1995): 9–10.
19. Tony Frudakis and others, "A Classifier for SNP-Based Racial Inference," Journal of Forensic Science 48
(July 2003): 771–82. For further information, see Tony N. Frudakis, "Powerful but Requiring Caution: Genetic
Tests of Ancestral Origins," in the present issue of the NGS Quarterly.
National Genealogical Society Quarterly 250
areas of the world is possible, but accuracy depends on the depth of sampling from
each region. Participants receive "likelihood scores" for each level, which enable
them to weigh the results. Assignment to broad ethnic and geographic classifications
applies to questions of deeply rooted ancestry. In contrast, inferring
membership in more localized populations can provide information on a genealogically
useful scale.20
DNA IN THE LABORATORY
Today commercial laboratories apply techniques that are spin-offs from the
Human Genome Project, the massive international collaboration to analyze the
complete set of human chromosomes.21 Nevertheless, current technology has not
yet advanced to the point where laboratories can report an individual's entire
genetic makeup.22 Instead, specialized tests analyze limited sections of DNA to
help solve problems in areas including crime investigation, paternity, identification
of human remains, and genealogy.
Mitochondrial DNA Sequences
DNA consists of long sequences of four chemical compounds. These building
blocks—called "bases" or "nucleotides"—are often abbreviated as A (adenine), C
(cytosine), G (guanine), and T (thymine). Scientists were initially skeptical that
such a limited set of chemicals could account for the complexity of life. However,
the four bases can be arranged in many different orders, just as letters from the
English alphabet can be shuffled in many different combinations and chained
together into a complete book. The human genome contains about three billion
bases. Some of its sections can be decoded, but a surprisingly large amount,
perhaps as much as 98 percent, appears to be meaningless. Genealogical tests
focus on these regions of "junk" DNA, and thus they cannot reveal any personal
traits or medical conditions. For an example of a short sequence of DNA bases,
see table 1.
A cell's mtDNA contains 16,569 sequenced bases but, for genealogical questions,
laboratories typically study segments containing only 400 to 1,100 of the
most informative bases. These sections are called "hypervariable" because they
show more differences among people than mtDNA's other regions. Because a
report listing even 400 bases would be difficult to interpret, laboratories conducting
mtDNA tests customarily report only the bases that differ from a standard
sequence called the Cambridge Reference Sequence (CRS). For example, a re-
20. Jayne E. Ekins and others, "Inference of Ancestry: Constructing Hierarchical Reference Populations
and Assigning Unknown Individuals," Human Genomics (forthcoming).
21. "The Human Genome Project Completion: Frequently Asked Questions," National Human Genome
Research Institute (http://www.genome.gov/11006943).
22. Corie Lok, in "Deciphering DNA, Top Speed," TechnologyReview.com (http://www.technologyreview
.com/articles/05/05/issue/forward_dna.asp?p=1), writes "Using about 100 state-of-the-art sequencing machines
to fully sequence the 3.2 billion DNA letters that make up one person's genome would take six months and cost
$20 million to $30 million."
The Science of Molecular Genealogy 251
port of an "HVR 1" test result as "16093C" would mean that the subject's
mtDNA in hypervariable region one (HVR 1)—an mtDNA segment containing
bases in positions numbered 16,024 through 16,365—differs from the CRS because
it has a cytosine (C) base at position 16,093. Most people have a few
differences from the CRS.
The Smallest Changes in DNA
As described above, mutations are modifications in DNA molecules that
occur randomly. Mutations can have positive or negative effects, but they typically
occur in sections of DNA that have no effect. A mutation that replaces just
one base (nucleotide) with another is called a "single nucleotide polymorphism"
or "SNP" (pronounced "snip"). For an example, see table 1.
The changes in the mtDNA molecule described above are literally SNPs, but
more often the term is applied to SNPs sprinkled throughout the chromosomes,
including the Y. SNPs, which tend to be rare, often represent unique events.
Given their low rate, SNPs are used in anthropological studies for tracing extremely
deeply rooted pedigrees, for example, determining matrilineal or patrilineal
descent from one of several ancient "clans."23
By using multiple SNPs researchers can determine the order in which the
SNPs occurred and estimate when two ancient lineages diverged. See table 2.
The variability of SNPs among descendants of a "founding father" gives a rough
estimate of when he lived: the more SNPs the descendants have, the more time
has elapsed since their lineages diverged. For example, The Y Chromosome
Consortium has identified a set of SNPs useful in classifying males into hierarchically
related clusters. Small clusters with identical haplotypes can be combined
into larger groups with similar but not identical haplotypes and so forth
23. Bryan Sykes, The Seven Daughters of Eve: The Science that Reveals our Genetic Ancestry (New York:
W. W. Norton, 2001).
National Genealogical Society Quarterly 252
until the groups are joined into a tree representing all humankind. The clusters,
labeled in a systematic alphanumeric fashion similar to an outline, comprise
haplogroups and subhaplogroups.24 Because different haplogroups predominate in
different regions of the world, Y-chromosome and mtDNA haplogroups suggest
ethnic and geographic origins of patrilineal and matrilineal ancestry—for example
a man in Y-chromosome group R1b might have male-line ethnic origins in
Western Europe.25
The Most Genealogically Useful Changes in DNA
The genetic information that genealogists most often employ is the "short
tandem repeat" (STR). The term refers to the repetition of a short sequence
of bases. For instance, a sequence of four bases, like G-A-T-A, might occur
seven consecutive times in one segment of DNA. When an STR mutates, the
number of repetitions changes—for example seven repetitions of G-A-T-A at
24. "A Nomenclature System for the Tree of Human Y-Chromosomal Binary Haplogroups," The Y Chromosome
Consortium (http://ycc.biosci.arizona.edu/nomenclature_system/frontpage.html).
25. J. Douglas McDonald, "World Haplogroup Maps," McDonald Group (http://www.scs.uiuc.edu/
∼mcdonald/WorldHaplogroupsMaps.pdf).
The Science of Molecular Genealogy 253
one location on a father's Y chromosome might change to six repetitions of the
same sequence at the same location on his son's Y chromosome. For an example
see table 3. Such mutations pass unchanged from parent to child until another
mutation occurs.
As various distinct STR mutations accumulated in different lineages over
many centuries, each developed its own pattern of STRs. Consequently, people
with different lineages have distinct inherited STR patterns. Individuals with
identical patterns are said to bear the same "haplotype," such as the Jefferson
Y-chromosome haplotype of Eston Hemings's male-line descendants. Minor differences
in haplotypes are compatible with descent from a common ancestor.
Genetic tests determine the pattern of STRs on the Y chromosome. Locations
or segments on the chromosome often have labels starting with the letters DYS
(for example, "DYS439"), which stand for "DNA Y-chromosome sequence."26
Genetics laboratories test between twelve and forty locations on the Y chromosome.
Their reports list the DYS numbers of the locations tested and the number
of STRs in each location—such as 13 at DYS 393. For example, the DNA test
that determined the Jefferson haplotype reported the number of STRs at eleven
DYS locations on a Jefferson descendant's Y chromosome. See table 4.
DNA ON THE INTERNET
Y-chromosome Databases
Most Y-chromosome tests take place on a small scale within surname projects,
such as the Edmund Rice study described above. However, large assemblies of
Y-chromosome data are available on the Internet, which can place an individual's
test results in a global context. Such public databases might generate privacy
26. John M. Butler, Forensic DNA Typing: Biology, Technology, and Genetics of STR Markers (Burlington,
Mass.: Elsevier Academic Press, 2005), 23–25.
National Genealogical Society Quarterly 254
concerns, but they do not reveal personal identities, and their data do not
contain information about personal traits or medical conditions. The rapidly
accumulating volume of online genetic information can aid investigations of
genealogical questions.
Results from the 1998 study of Carr, Jefferson, and Woodson haplotypes
illustrate the useful information that can be gleaned from online databases. A
centerpiece argument was the rarity of the Jefferson haplotype, suggesting that
the Jefferson-Hemings match was not coincidental. When the article was written,
the database of Y-chromosome haplotypes contained 670 European records,
with no matches to the Jefferson haplotype.27 Today, however, such databases
have vastly larger numbers of records:
• Y-chromosome Haplotype Reference Database (YHRD) is an anonymous database
of records submitted by forensic laboratories and collected to provide a cross-sample
of people in specific locations. It contains 28,650 world-wide records, including
18,711 from Europe.28
• Ybase was the first publicly accessible database that allowed individuals who had
used different testing companies to enter their data and compare results. It contains
5,025 Y-chromosome haplotypes, 6,214 surnames, and useful statistical summaries
showing the range of STRs for tested Y-chromosome locations.29
27. Foster and others, "Jefferson Fathered Slave's Last Child," 27–28.
28. "About the 'YHRD - Y Chromosome Haplotype Reference Database'," YHRD.org (http://www.yhrd
.org).
29. Ybase: Genealogy by Numbers (http://www.ybase.org).
The Science of Molecular Genealogy 255
• Ysearch is a publicly accessible database containing approximately thirteen thousand
Y-chromosome records with the large majority having test results for twelve to
twenty-four markers. Anyone can manually add data obtained from any company,
and an automated procedure is available for Family Tree DNA customers. Users can
add genealogical information to their Y-chromosome records.30
• The Sorenson Molecular Genealogy Foundation (SMGF) database includes 13,489
Y chromosomes linked to 550,000 ancestors.31 With 9,400 unique surnames and
more than 90 percent of the Y-chromosome haplotypes tested at thirty or more
markers, it is the largest searchable Y-chromosome database. SMGF analyzes
samples contributed by volunteers, who can order a free participation kit from the
Web site.32
• Ymatch contains thirty-five hundred records with the majority having test results
for twenty-six to forty-three Y-chromosome locations. This is the newest database
of this kind available to the public.33
None of the above sources contains a match for the Jefferson haplotype. In
contrast, the Sorenson database has 531 records, with many different surnames,
matching the Carr haplotype shown in table 4, which was based on eleven
markers (seven markers, when applying modern standards). Testing more markers
on the Carr sample, as various companies do today, would reduce the matches to
those most closely related to the Carrs who were tested.
The results for the Woodson haplotype, depicted in table 4, are instructive.
The Sorenson database has just one match, a sample from Ghana. That does not
necessarily mean that the Woodson line came from Ghana—a larger database
could show matches in other localities. However, this result suggests that the
haplotype is more typical of African ancestry than European. Such geographical
information may shed light on ancestral lines that lack documentary evidence for
their origins.
Mitochondrial DNA Databases
Just as with Y-chromosome analysis, a person with mtDNA test results can
compare them with various online databases. DNA data for the "Ice Man"
demonstrate the information available. A body discovered in 1991 at the edge of
a melting Alpine glacier was initially thought to be a climber who had died in
modern times. Scientists soon determined, however, that the remains were some
five thousand years old.34 The Ice Man had a relatively common mtDNA haplotype
present in 1 to 2 percent of Europeans. Containing two differences from
30. Ysearch (http://www.ysearch.org).
31. Ugo A. Perego, Natalie M. Myres, and Scott R. Woodward, "'Y' Research Through DNA," Everton's
Family History Magazine 58 (May–June 2004): 26–28.
32. Sorenson Molecular Genealogy Foundation (http://www.smgf.org).
33. Relative Genetics (http://www.relativegenetics.com).
34. Oliva Handt and others, "Molecular genetic analyses of the Tyrolean Ice Man," Science 264 (17 June
1994): 1775–78.
National Genealogical Society Quarterly 256
the Cambridge Reference Sequence (CRS), his sample had cytosine (C) bases at
positions 16,224 and 16,311 in hypervariable region 1.35
The following online databases contain mtDNA test results:
• MitoMap lists positions where differences from the CRS have been reported. These
are listed by mtDNA location number rather than as a composite haplotype. However,
it can be seen that many studies have reported differences. Occasional individuals
have a novel difference, one never previously described.36
• The Mitochondrial DNA Concordance is a collection of several thousand mtDNA
haplotypes reported in the technical literature up to about 1998.37 The Ice Man's
mtDNA haplotype (16224[C] 16311[C]) can be found in two places, under
the listings for locations 16,224 or 16,311. The database shows that the Ice
Man's haplotype occurs in many European populations—some Basque, Bavarian,
Bulgarian, Cornish, English, Finnish, German, Norwegian, Portuguese, Swiss,
Tuscan, and Welsh people have the same haplotype. Thus it is not possible to
ascribe the Ice Man's ancestral line to a specific European location; however,
because the haplotype is not found on other continents, the broad classification of
European ancestry is confirmed.
• Oxford Ancestors offers guest access to its database. Entering the Ice Man's mtDNA
test results shows 250 matches. Oxford Ancestors classifies him as a member of the
"Katrine" clan, a pseudonym for mitochondrial haplogroup K.38
• Mitosearch is a public-access database of individually contributed GEDCOM files
and mtDNA test results that have not been independently verified. A recent survey
of the database produced fifty-eight matches for the Ice Man. It also yielded 150
members of haplogroup K. They had the Ice Man's two differences from the CRS
plus various additions, demonstrating variation within a haplogroup.39
• The mtDNA Log functions like a guest book. Contributors may leave free-form
comments along with their mtDNA test results. The site does not have a search
function, but the Web browser's "Find" function substitutes. Visitors often provide
data about their ancestral names and geographical locations.40
• The Federal Bureau of Investigation maintains an "mtDNA Population Database,"
which incorporates sequences from the Mitochondrial DNA Concordance (above)
as well as more recent contributions from accredited forensic testing laboratories.
This anonymous database can reveal whether a haplotype is common or rare.41
35. Michael D. Coble and others, "Single Nucleotide Polymorphisms Over the Entire mtDNA Genome
that Increase the Power of Forensic Testing in Caucasians," International Journal of Legal Medicine 118 (June
2004): 137–46.
36. MITOMAP: MtDNA Control Region Sequence Polymorphisms (http://www.mitomap.org/cgi-bin/
mitomap/tbl6gen.pl : dated 27 September 2005).
37. Kevin Miller and John Dawson, Mitochondrial DNA Concordance (http://www.bioanth.cam.ac.uk/
mtDNA/).
38. "Oxford Ancestors' Databases," Oxford Ancestors (http://oxfordancestors.com/members).
39. Mitosearch (http://www.mitosearch.org).
40. Charles F. Kerchner Jr., Mitochondria DNA (mtDNA) Test Results Log (BLOG) (http://www.kerchner
.com/cgi-kerchner/mtdna.cgi).
41. Keith L. Monson and others, "The mtDNA Population Database: An Integrated Software and Database
Resource for Forensic Comparison," Forensic Science Communications 4 (April 2002), electronic edition
(http://www.fbi.gov/hq/lab/fsc/backissu/april2002/miller1.htm).
The Science of Molecular Genealogy 257
• GenBank is a repository at the National Institutes of Health for raw mtDNA
sequences from technical literature. With some effort public users can align their
sequences to the published data and compare the results.42
• Many individuals have developed custom Web sites devoted to a specific haplogroup.
The World Families Network maintains a page of links to these special
interest groups.43 For instance, John Walden's Web site for haplogroup K diagrams
the relationships between the "clan mother" for haplogroup K and the variations he
has located. 44
DNA TESTING IN THE FUTURE
In years to come DNA tests probably will be faster and cheaper and will
include more markers than today's tests. Publicly accessible databases of compiled
genetic information will also continue to grow, allowing genealogists to correlate
DNA test results with population-based studies, such as the National Geographic
Society's recently launched Genographic Project.45 Large databases containing
mtDNA and Y-chromosome matches could suggest research pathways that might
unblock a lineage problem that seems unsolvable with documentary research
alone.
The bulk of current tests for genealogical purposes is limited to the Y chromosome
and mtDNA. This is a severe constraint because straight paternal and
maternal lineages represent only a tiny fraction of anyone's total ancestry and
DNA. The other parts of the pedigree harbor vast amounts of information that
future genetic testing might unlock. Laboratories are beginning to study the use
of autosomal DNA for genealogical purposes.
CONCLUSION
Molecular genealogy synthesizes traditional genealogical research and relatively
new technologies developed to explore genetic characteristics of the
world's people. The combination enhances traditional genealogical methods,
especially when ambiguities and roadblocks in written records impede documentary
research. Scientific methods are just beginning to tap into the invaluable
repository of ancestral information that is carried in every individual's DNA.
Molecular methods can help individuals uncover previously unknown family
relationships, verify or refute claims to ancestry, and shed light on questions that
have puzzled genealogists for years.
Currently the two most active areas of genetic testing for genealogical purposes
focus on mtDNA and the Y chromosome. DNA projects for family history
42. "GenBank Overview," National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/
Genbank/).
43. "mtDNA—The Family of Woman," World Families Network (http://worldfamilies.net/mtDNA.htm).
44. John S. Walden, "Swinging in the mtDNA Tree," Walden/Adams/Walts/Waltz Surname DNA Projects
(http://freepages.genealogy.rootsweb.com/∼jswdna/mtdna.html).
45. NationalGeographic.com, The Genographic Project (https://www3.nationalgeographic.com/
genographic).
National Genealogical Society Quarterly 258
purposes can use samples from only two participants or hundreds. Many questions
can be approached by querying online searchable genetic databases to find genetic
matches to a known DNA profile. These remarkable resources are freely
available and continuously expanding. Molecular genealogy methods eventually
will enable genealogists to explore lines beyond strictly matrilineal and patrilineal
ancestry. In the near future genealogists can expect a burgeoning expansion
of this field. Genetic testing will be more widely available, increasingly economical
for the individual, and more informative for answering a greater variety of
genealogical questions.
Emigrated Before Birth?
[William Spreen declaration of intent, Taylor County, Wisconsin, Declarations
of Intention 4: 408, Wisconsin Historical Society, Madison; microfilm 2,134,516,
Family History Library, Salt Lake City, Utah. Underlined portions are handwritten
on a printed form.]
I, William Spreen, aged 28 years, occupation Farmer, do declare on oath that my
personal description is: Color White, complexion Light, height 5 feet 7 inches,
weight 145 pounds, color of hair Brown, color of eyes Blue[,] other visible distinctive
marks None; I was born in New Stetten, Germany, on the 19 day of
January, anno Domini 1890; I now reside at Medford Wisconsin. I emigrated to
the United States of America from Old Stetten, Germany on the vessel Unknown;
my last foreign residence was New Stetten, Germany. It is my bona fide
intention to renounce forever all allegiance and fidelity to any foreign prince,
potentate, state, or sovereignty, and particularly to William II German Emperor,
of which I am now a subject; I arrived at the port of New York, in the State of
New York on or about the 15 day of June, anno Domini 1889; I am not an
anarchist; I am not a polygamist nor a believer in the practice of polygamy; and
it is my intention in good faith to become a citizen of the United States of
America and to permanently reside therein: So help me God. William Spreen
(original signature of declarant.) Subscribed and sworn to before me this 8th day
of January, anno Domini 1919. (seal) S. A. McComber, Clerk of the Circuit
Court.
—Contributed by Joy Reisinger, CG
The Science of Molecular Genealogy 259
