Solving crime means finding out whodunit, so forensic scientists have long searched for ways of absolutely identifying individuals from materials left at a crime scene.
The first discovery that provided positive proof was fingerprints, which are absolutely individual. No two people share the same prints, so fingerprinting became and remains an extremely powerful forensic tool. However, fingerprints aren't found at every crime scene. Criminals have learned to wear gloves and to wipe their prints from any objects they touch.
Tracking down every bit of biological debris that gets left behind, however, is impossible for even the best criminals. DNA fingerprinting gives the criminalist a relatively new and extremely accurate tool for using the tiniest bits of genetic material to identify individuals who were present at a crime scene.
Tracking down and preserving DNA
DNA is found in almost every cell in the human body. Skin, hair follicles, semen, saliva, and blood are common sources of crime-scene DNA. Hair doesn't contain cells, but hair follicles do. Saliva doesn't contain cells, but as it passes through the salivary ducts and washes around the mouth, it picks up cells from the ducts and the mouth. RBCs have no nuclei, so they contain no DNA. The DNA found when blood is tested comes from the white blood cells (WBCs). Using modern techniques, each type of fluid or tissue yields enough usable DNA for testing.
After it's secured from a crime scene, DNA must be handled carefully to keep it from degrading. The best DNA samples are the ones that have been adequately dried and stored in protective containers. When drying isn't feasible, wet samples need to be frozen until they're analyzed.
Imagine trying to read a book in which all the sentences had been reduced to fragments or single words. War and Peace might be indistinguishable from Green Eggs and Ham. However, if the original books were merely torn into pages, you'd have little trouble distinguishing between the two. Similarly, DNA typing and matching depends upon the preservation of the sequence of the bases that make up the DNA. If the lab has only very short fragments or single bases to work with, it can't effectively type the DNA.
The bigger the DNA sample, the better, and yet usable DNA has been obtained from small and unlikely sources. Even a toothbrush, stamp, or bite wound can yield a usable saliva sample. A single drop of blood or a single hair follicle often is enough.
In addition, modern technology makes possible the extraction of usable DNA from ancient tissues, even those taken from mummies that are thousands of years old. Scientists have extracted DNA from the bones and teeth of very old skeletal remains, and at times, from severely burned bodies.
Looking into the genome
The genome is the total DNA within the cell, or the millions of base pairs that make up the long polymers of DNA. Out of that massive number of base pairs, only about 5 percent directly carry out the work of life. These genes are encoded, meaning that they direct the synthesis (or manufacture) of proteins that the body needs for growth and function. The other 95 percent of the genome is non-encoded, which means it doesn't directly code for the production of a protein, but it doesn't simply lie around doing nothing, either. A portion of it regulates how genes function, and much of it is repetitive information whose purpose scientists haven't yet been able to identify.
All humans, and indeed all primates, share a large amount of the genome, meaning that much of your DNA is exactly like mine and everyone else's. It's also identical to that of the chimpanzees at your local zoo. Even so, that leaves plenty of unique combinations of DNA to give forensic investigators a keen method of finding out exactly who you are.
In 1985, Alec Jeffreys and his associates at Leicester University discovered that each person's DNA is actually unique. They found that certain areas of the long human DNA molecule exhibit polymorphism, a fancy word that means it can take many different forms. These variable areas are unique in everyone, and analyzing these areas allows scientists to make distinctions between one individual and the next. Shortly after discovering this polymorphism, Jeffreys developed a process for isolating and analyzing these areas of human DNA that he termed DNA fingerprinting. Currently, the process is also called DNA typing.
Polymorphisms important for forensics can be found in non-encoded, or junk DNA. These areas are highly variable in length and base sequence. The variability in length is called length polymorphism. It's an important factor in forensic DNA typing because certain base sequences within the non-encoded DNA segments are constantly repeated. As a result, forensic investigators look for two types of sequences:
- Variable Number Tandem Repeats (VNTRs): The same base sequence repeats throughout a specific locus within the strand. These segments can be hundreds of base pairs long, repeating along the length of the DNA strand a variable number of times.
- Short Tandem Repeats (STRs): Much shorter than VNTRs — usually three to seven base pairs long — these sections also repeat throughout portions (loci) of the DNA chain. STRs repeat over segments of the DNA strand as long as 400 bases, which means that by using STRs, lab technicians can use even severely degraded samples for testing. Many more STRs are known than VNTRs, which gives forensic scientists many more repeats to analyze.
The key in DNA typing is that the variability in the pattern of these repeats from person to person is broad, meaning that if technicians can isolate a certain locus of the DNA strand and determine the number of repeats of a given sequence in that area, they can compare it with another DNA strand to find out whether the pattern matches. In addition, research has determined how often a given number of repeats is found at a specific location in the DNA of the general population. Criminalists can use that information to calculate the probability that two DNA samples came from the same person. However, a match from a single locus is not very conclusive. But if several loci match, the probability quickly adds up.
Repeating yourself: How duplication identifies you
In terms of DNA, all of us repeat ourselves, but the specific ways in which we do it make each of us unique. When working to match DNA, investigators look at the repeats on particular loci of DNA. If my DNA were being compared to yours, for example, investigators would look at the same locus on each of our samples. They may find that you received 8 repeats of a particular STR from one parent and 14 repeats from another, and that I received 15 repeats of the same STR from one parent and 23 from another. Your DNA and mine would be very different.
But would your and my DNA be different from everyone else's on Earth? You couldn't tell by looking at only one locus. Other people also may have received 8 and 14 or 15 and 23 repeats for the same locus, but when you look at a dozen loci, the probability that two people received the exact number of repeats from each parent at all 12 loci is only one in several hundred trillion.
Looking at another example, when you analyze the STRs of a crime-scene sample at five different loci, you may find the following repeats:
Locus 1 12 and 9
Locus 2 6 and 14
Locus 3 23 and 16
Locus 4 5 and 18
Locus 5 8 and 19
Now, say that you already know that each of these STR repeat patterns occurs at these specific loci at respective rates of 1 percent, 3 percent, 8 percent, 1 percent, and 2 percent within the general population. That means 1 in 100 people share the same repeat pattern at Locus 1, 3 in 100 share this same repeat pattern at Locus 2, and so on. Therefore, if a suspect's DNA shows the exact same repeat patterns at all five loci as the crime-scene sample, the probability that the DNA found at the scene came from someone other than the suspect is tiny. In fact, because the inheritance of the STR patterns at each locus is independent of any other locus, the percentages must be multiplied by each other to determine the probability of the DNA coming from someone other than the suspect, which in this case is a whopping 48 out of 10 billion, and that high degree of probability was found using only five loci. Imagine what those odds would be if the suspect's DNA matched the crime-scene sample at 12 or more loci. Book him, Dano.