Paternity Testing: Blood Types and DNA

Occasionally, situations arise in which people require concrete, scientific evidence of parentage, whether it be their own or that of someone else. In most instances, maternity is easy to determine. Before surrogate motherhood became possible, the woman who gave birth to a child was obviously that child's gestational, genetic, and legal mother, and this continues to be true in the vast majority of cases today.

Unfortunately, questions of paternity aren't so easy to answer. In order to make a determination of fatherhood, scientists almost always work backwards--from the child to the potential parent--to ascertain the actual nature of the relationship. In the past, this typically involved identifying specific phenotypes (in particular, specific blood types) in the child and using this information to either "rule in" or "rule out" possible fathers. However, this system presented a number of problems, not the least of which was that it often yielded inconclusive results. Thus, since the 1990s, the more common approach has been to consider the presence of particular genotypic markers when attempting to establish fatherhood (and, in a handful of cases, motherhood).

The process of DNA fingerprinting was developed by Alec Jeffreys in 1984, and it first became available for paternity testing in 1988. Before this sort of DNA analysis was available, blood types were the most common factor considered in human paternity testing. Blood groups are a popular example of Mendelian genetics at work. After all, there are numerous human blood groups with multiple alleles, and these alleles exhibit a range of dominance patterns.

Today, the best-known blood-typing system is ABO typing, which involves the presence of antigens on red blood cells that are encoded by the ABO locus on human chromosome 9. In the ABO system, the A allele and the B allele are codominant, and the O allele is recessive. Thus, if a person's ABO blood type is O, he or she has two O alleles. If, however, a person's blood type is A, he or she has either two A alleles or one A allele and one O allele. Similarly, if a person has type B blood, this indicates the presence of either two B alleles or one B allele and one O allele. Finally, some people have type AB blood, which means they inherited both an A allele and a B allele.

In cases of questioned paternity, ABO blood-typing can be used to exclude a man from being a child's father. For example, a man who has type AB blood could not father a child with type O blood, because he would pass on either the A or the B allele to all of his offspring. Despite their usefulness in this regard, ABO blood groups cannot be used to confirm whether a man is indeed a child's father. Because of this and several other factors, it took the legal system some time to trust blood-typing. For example, in a famous case in 1943, the starlet Joan Barry accused actor Charlie Chaplin of fathering her child. Although blood tests definitively excluded Chaplin as the father, the court did not allow this evidence to be admitted, and Chaplin was ordered to pay child support to Barry. The Barry/Chaplin case did spur the passage of new laws, however, thus launching a new era in forensic evidence.

Over time, the use of additional blood antigens, such as those associated with the MN and Rh systems, refined the use of blood-typing for both paternity and forensics. However, such blood groups were only about 40% effective in ruling out a man as a child's father. Then, in the 1970s, testing for human leukocyte antigens (HLAs) added a distinguishing feature that made it possible to rule out men as fathers with 80% effectiveness. The genes responsible for the HLA system are involved in antigen presentation to T cells. The HLA system is highly polymorphic, with more than 3,200 different alleles identified so far (Robinson et al., 2003; Williams, 2001). Although this vast number of alleles causes headaches for cell and organ transplants, the multiplicity of genotypes the HLA system provides—in the tens of millions—makes it ideal for consideration in identity and paternity testing.

Figure 1: Paternity testing using microsatellite markers.

This test includes samples from the mother (top row), the child (middle row), and the alleged father (bottom row). The maternal marker that has been passed to the child is 6. This means that the other marker present for the child (7) must have been inherited from the father. The alleged father matches the child, since one of his markers is indeed 7.

© 2008 Gendia All rights reserved.

In the 1970s and 1980s, electrophoresis of various biochemical markers became widely available. With this process, proteins from a person's blood or other tissue were placed onto a gel, such as potato starch, agarose, or polyacrylamide. An electric current was then run through the gel, and different forms or isozymes of the proteins were separated by their electrical charge and/or size. Differences in isozymes relate to differences in the alleles that code for these proteins. Thus, the presence of certain identical isozymes in samples from both a child and his or her potential father could be used to reveal the existence of a genetic relationship between the two individuals (Figure 1). In fact, by 1974, Chakraborty et al. suggested that genetic testing via electrophoresis had advanced such that this method might be used to confirm paternity rather than merely exclude a man as a child's father.

Today, with the advent of numerous DNA sequencing, amplification, and testing techniques, paternity testing has evolved even further than predicted. Indeed, present-day genetic testing has an accuracy rate of up to 99.99% (i.e., 9,999 out of 10,000). Of course, the exact level of accuracy depends on the number and quality of the genetic markers being considered. (Here, it is important to emphasize that scientists consider only specific marker alleles, rather than entire genomes, when conducting paternity testing. Full genome analysis would add a great deal of time and expense to the process without significantly improving the accuracy of the results.) Thus, DNA-based forms of paternity testing have all but taken over earlier methods. In addition, higher throughput, better sensitivity, and automation have allowed DNA testing to be performed on ever-smaller and sometimes degraded DNA samples with greater speed and excellent accuracy.

Interestingly, improvements in paternity testing over the past several decades have not only led to an increase in the accuracy of test results, but also to expanded application of various testing methods. For example, as DNA technology has gotten more precise, it has become possible to determine paternity using DNA from grandparents, cousins, or even saliva left on a discarded coffee cup. Such DNA testing is clearly an important part of criminal investigations, including forensic analysis, but it is also useful in civil courts when the paternity of a child is in question. The widespread availability of paternity tests on the web and in neighborhood drugstores is also indicative of a civil demand for this technology. However, it is important to note that such direct-to-consumer (DTC) tests will not stand up in court because there is no way to prove whose samples were analyzed. Hence, DTC testing is most often used to assist in making a decision to initiate litigation or to simply provide peace of mind in matters of questionable paternity.

In broader applications, advances in paternity testing mean that people who were adopted now have more direct means to confirm their biological identity or to find their birth parents. In addition, parentage testing is often an essential tool in proving immigration status in cases of family reunification.

For years, questions of paternity presented a significant challenge to scientists and potential parents alike. During the first half of the twentieth century, researchers often turned to people's ABO phenotypes when such issues arose; however, ABO blood group information could only be used to exclude potential fathers, rather than confirm the presence of a parental relationship. Consideration of additional blood markers, such as Rh antigens, MN antigens, and HLAs, greatly increased the effectiveness of paternity testing over the next few decades, but it still left significant room for error. Thus, with the dawn of DNA analysis and sequencing techniques in the 1980s and 1990s, scientists increasingly began to look at people's genomes when questions of fatherhood arose. This approach proved exceedingly useful; in fact, current marker-based methods of analysis yield test results that are both 99.99% accurate and applicable in a variety of settings. With the ongoing advancement of DNA sequencing and analytical technologies, we will no doubt continue to see an increase in the utility of these tests, as well as in the availability of detailed genetic services to the general public.