A genetic marker is a trait transmitted from parent to child, thus potentially permitting the reconstruction of patterns of descent on the basis of its distribution among population members. Genetic markers are thus estimators of biological relatedness, and when they are shared by individuals they are interpreted as evidence of a “natural” link between them. Genetic markers have been used to establish paternity, to identify the origin of a biological sample at a crime scene, to create maps of disease-causing genes, and to link people who have never met one another into new networks of kinship relations. The reliability of the inferences derived from any genetic marker is a function of three properties: (1) mode of inheritance, (2) stability, and (3) rarity.
The mode of inheritance of a genetic marker may be quite variable. Surnames in many societies are inherited in a fashion that mimics a Y chromosome (that is, a father and son are very highly correlated), and thus can serve as a noninvasive estimator of relatedness. (The method of tracking genetic inbreeding in a population through surname distribution is called isonymy.) By contrast, chromosomes from the cell’s nucleus are transmitted to offspring biologically. However, because each parent passes on only one member of each pair of chromosomes, a child ordinarily has only a 50 percent chance of matching a parent’s (or sibling’s) corresponding genetic marker. There are exceptions to this rule:. Most of the Y chromosome is transmitted intact from father to son, so that the Y chromosomes of a father and son should be nearly identical.
Likewise, the DNA of the mitochondria, which exist in the cell but outside the nucleus, is transmitted intact from mother to offspring, so that there should be a perfect match between a mother and child (and the father is paradoxically unrelated to the child with this marker).
Stable genetic markers are preferred for comparisons because they enable tracking over many generations. Physical features with multifactorial or polygenic causes, such as features of the bones and teeth, may appear to blend away over generations from intermarriage, or they may have their expression altered by the environmental conditions in which the organism grows and develops. Very rapidly mutating segments of DNA may be just as compromised for use as genetic markers, if their rate and mode of change preclude a secure match between related individuals.
A common genetic marker is less valuable than a rare one, for the simple reason that a match between two samples is more likely to be due to chance, rather than to familial descent, if the genetic marker is a common one. Because type O blood is the most ubiquitous blood type among all human populations, two people who are not close relatives are nevertheless very likely to exhibit this genetic marker. It would consequently be a genetic match, but not a very informative one.
Even before the development of the science of genetics, however, similarities of the language, skeleton, and teeth were being understood as crude genetic markers. Certain traits (ranging from diseases or deformities to simple quirks) were recognized to run in families, and thus to attest to close kinship among the bearers of such traits. In the early part of the twentieth century, serologists developed blood tests to detect biochemical differences among people, which were very close to direct products of the genes, if not the genes themselves. The most immediate value of these differences was in paternity exclusion, but they were also quickly adopted to study racial relationships. This proved to be a very frustrating exercise because groups of humans identified serologically corresponded very poorly to common concepts of “races” (see Marks 1995).
In modern forensic contexts, in order to connect a suspect to a sample (and to rule out the possibility of such a match coming at random), genetic markers need to be both highly variable and individually uncommon. Short, localized repetitive DNA sequences, in which the number of tandem repetitions of the specific DNA sequence varies strikingly from person to person, have proven to be most effective for this purpose. In gene-mapping studies, DNA markers are commonly differences of a single base among individuals in a population (such differences are called single nucleotide polymorphisms, or SNPs). A linear series of these genetic markers, usually transmitted as a single block of DNA, is called a haplotype.
Genetic ancestry services principally use markers derived from Y chromosome DNA and mitochondrial DNA to establish matches between samples of presumptive relatives. Others use nuclear DNA markers from small samples of diverse peoples as a baseline to establish a customer’s “racial” affiliation, which simply expresses an overall pattern of similarity to one or more of these standard samples.
Bolnick, Deborah A. Forthcoming 2008. “Individual Ancestry Inference and the Reification of Race as a Biological Phenomenon.” In Revisiting Race in the Age of Genomics, edited by Barbara Koenig, Sandra Lee, and Sarah Richardson. Piscataway, NJ: Rutgers University Press.
Marks, Jonathan. 1996. “The Legacy of Serological Studies in American Physical Anthropology.” History and Philosophy of the Life Sciences 18 (3): 345–362.