DNA, or deoxyribonucleic acid, is often called the genetic blueprint of an individual. Just as a carpenter or builder uses an architect’s blueprint to construct a home or office, the body uses DNA as its blueprint to construct an individual. With the exception of identical twins, no two people are exactly alike and therefore, no two people’s DNA is exactly alike either. Furthermore, we all inherit half of our DNA from our mother and half from our father. These two facts about DNA are the basis for DNA testing for criminal investigations and parentage testing.
Each cell in our body contains the same DNA (except germ cells, sperm, and eggs), which is why DNA testing can be performed from blood, oral (mouth) swabs or any other biological specimen that carries nucleated cells (cells with a nucleus). Mature red blood bells (RBC) are anucleate (no nucleus) and therefore, they don’t carry DNA, at least not the kind we usually test.
The DNA in people is packaged into structures called chromosomes. Chromosomes are made up largely of protein with the DNA wrapped around the protein structures. Humans have 23 pairs of chromosomes for a total of 46 chromosomes. One chromosome of each pair comes from the mother and one comes from the father. The chromosomes are found within the nucleus of the cell.
If you were to unwind the DNA from each of the 46 chromosomes from a single cell and stretch it out end to end it would measure about 6 feet—each microscopic cell in your body has about 6 feet of DNA packed inside of it. The average person is composed of about 100 trillion cells and each of those contains about 6 feet of DNA. If you took the 6 feet of DNA from each of the 100 trillion cells in a body and stretched it end to end, it would measure some 125 billion miles, or about 1000 times the distance between the earth and our sun.
But much of this DNA is the same between any two people. Most of it is pretty much the same from person to person with slight differences to give some people dark hair and skin and others light hair and skin and still others something in between. But there are some regions of DNA that are highly variable from person to person. These regions largely lie in between genes and have traditionally been referred to as “junk” DNA. In the 1980s these variable regions of DNA were recognized to be composed of short regions of DNA that existed in tandem (head to tail, one after another like tandem axles on a trailer truck). What made them variable from person to person was the number of times the short region of DNA (also referred to as a DNA sequence) was present in any given person. In one person it might be present 11 times and in another 15 times and yet another perhaps 13 times. These regions of variable DNA positioned in tandem became know as Variable Number of Tandem Repeats or VNTR. More recently (in the 1990s) a related type of DNA started to be used for parentage and forensic testing that also represented a tandem repeat sequence but of a shorter length and came to be called short tandem repeats or STR.
In DNA testing for parentage, family studies, and criminal investigations, we zero in on those STR regions to identify people and determine what relationship exists, if any. These STR regions are located throughout our DNA on all of our 23 chromosome pairs. We inherit one chromosome of a pair from our mother and one from our father for a total of 23 pairs or 46 chromosomes. Any given STR on a chromosome has a corresponding STR region on its pair. Also, remember that these STRs are variable in size from person to person, or more accurately from chromosome to chromosome. So for any given STR the copy you inherit from your mother may be say 9 repeats in size and the one you inherit from your father may have say 12 repeats. Your DNA type at that STR would be designated 9,12.
Some of the most common STRs used in parentage and forensic testing are D3S1358, vWA, FGA, D8S1179, D21S11, D18S51, D5S818, D13S317, D7S820, D16S539, CSF1P0, TH01, TPOX, D19S433 and D2S1338. Of course, there are others too. These STR regions are named for their location in the genome (location on your chromosomes). For example, D21S11 refers to chromosome pair number 21 and the number 11 refers to a specific location on chromosome number 21.
A typical paternity report looks like this: Sample Paternity Report
And, a typical forensic report might look something like this: Sample Forensic Report
Now you know exactly what all those numbers mean. The column on the left lists the STRs that were tested (D8S1179, D21S11 etc.) and the numbers to the right of each STR refers to the DNA type of the individual.
Parentage/Relationship DNA Tests
In a routine paternity test we look at the mother’s DNA profile and compare her with the child to identify what the child inherited from the mother. For example, in the D8 location in the paternity test above, the mother and child share a 15. Therefore, the mother must have passed a 15 to her child. The remaining 13 in the child must therefore have come from the biological father. If the man being tested does not demonstrate a 13, then that would represent exclusionary evidence. Since biology is not quite so cut and dry it does take more than just one non-match or exclusion (also referred to as an inconsistency) to declare non-paternity. If the man being tested does demonstrate a 13 then that would represent inclusionary evidence or evidence in favor of paternity. We would conduct a similar analysis of the remaining STRs. If multiple STRs excluded the tested man then we would report that the tested man is not the biological father of the child. If on the other hand, the analysis did not exclude the man (the man matches the child) at any of the tested STRs then a statistical analysis would be performed to calculate the paternity index (genetic odds in favor of paternity) and the probability of paternity. The numbers in the far right column under Paternity Index represent the weight of the evidence for each locus or test system. For D8S1179 the tested man (John Doe) matches the child and the paternity index is 1.81. Since all of these test systems are independent of one another, the individual paternity indexes (or indices) are multiplied together to obtain the Cumulative Paternity Index (CPI) given in the body of the report (7,540,000). The product of the individual indices may not be exactly equal to the CPI given in the body of the report. This is due to something called significant figures and the rounding of numbers, but it will be very close. The CPI of 7,540,000 is the genetic odds in favor of paternity. Given the genetic data in the test, the tested man (John Doe) is 7,540,000 times more likely to father a child with the DNA profile than a unrelated and randomly selected man from the same population group (in this case White). Converting this to a probability of paternity yields 99.99999%.
If the parents are not available, testing two individuals to determine whether they are siblings can give insight as to whether the two might be related. The tests for sibling relationships depend on finding a sufficient number of genetic markers where the alleged sibs do share, and the sharing is for characteristics that are relatively uncommon between unrelated people. The test results are reported as a probability that the tested people are in fact siblings. That probability may be very high or quite low but it will not be zero. The acceptable level of probability may be at the discretion of the legal body that is requesting the test result.
In a full sibling test (i.e., with the same mother and father), If one parent is available for testing but the other is not, testing that available parent will be very useful in determining whether the other (alleged) parent could be the true parent of both children. If a half sibling test is needed (i.e., a test where two individuals may be related through only one parent) and the common parent is unavailable, testing the known parent of at least one of the siblings will allow the laboratory to see which of that child’s traits were actually conveyed from that parent. By default, that child’s other traits at the tested genetic loci must have come from the other parent. If enough of these “other parent” traits are shared and if they are sufficiently rare in the population as a whole, a convincing probability of a half sibling relationship may be calculated.
Forensics DNA Testing
In the sample forensic case reported above, police are investigating an assault of a women walking in a park. Witnesses stated that a man hit the victim and there was a struggle. Police recovered the victim’s shirt which had a stain that appeared to be blood. They also recovered the suspect’s shirt which also had a blood colored stain. The police submitted these items to the laboratory along with a reference sample from the victim (Jane Doe) and the suspect (John Doe).
The laboratory processed the evidence (the victim and suspect shirts) and determined that the stains were in fact human blood. Therefore, the laboratory conducted DNA testing of the evidence and then the reference samples from the victim and suspect. In this case we compare the DNA profile obtained from the blood stain to that of the victim and the suspect. Page 1 of the report provides a list of the items submitted for testing along with the final conclusions in the case analysis. Page 2 of the report provides the DNA profiles obtained from the evidence and the victim and suspect reference samples.
A review of the DNA profiles on page 2 of the report indicates that the DNA obtained from the blood on the suspect’s shirt matches the victim and the DNA obtained from the blood on the victim’s shirt matches the suspect. This genetic evidence tends to support the victim’s story as well as that of the witnesses although there may be other explanations for how the victim’s blood got on the suspect’s shirt and how the suspect’s blood got on the victim’s shirt. That is for the trier of fact to consider, the judge and jury. Note that the frequency of the DNA profile of the victim and suspect is provided in the report; less than 1 in 999 trillion in the general population. Since there are only about 6 billion people in the world an expected frequency of 1 in 999 trillion means that the profile is indeed very rare and therefore the two matching DNA samples originated from the same person. The exception of course is a case of identical twins who would have identical DNA profiles.