How Does DNA Fingerprinting Work

How does DNA fingerprinting works? DNA fingerprinting is a reliable method for identifying people. How DNA fingerprinting works as to do with something called the HLA system (human leukocyte antigen system). The HLA is a system that codes for proteins on the surfaces of many cells, including white blood cells. HLA genotyping is used in transplants. With this system, people could match transplant genetically, so that a transplant from one person to another wouldn’t be rejected by the recipient.

There are four different HLA genes, labeled nicely A, B, C and D. HLA A has 23 alleles. HLA B has 47. HLA C has 8 and D has 23. A person could be, for example, A11, B16, C5 and D11. With more alleles is more likely that people would be different from one another, and that parents and children can be matched better.

There are a couple of problems with the HLA system, however. First, you need well preserved tissue or blood. That’s a little difficult sometimes. Second, HLA proteins are not always present in our cells. Third, a lot of mixtures of these genes go on when gametes are produced. Fourth, there are pretty common HLA alleles, and some extremely rare ones.

The Short Tandem Repeats

Human genome sequencing has revealed that the genome contains short sequences that are repeated many times in tandem. These are appropriately called Short Tandem Repeats (STR). For example, let’s consider the DNA sequence TCAT. Looking through the whole genome, there are different Short Tandem Repeats, and the repeat numbers are inherited. You might inherit one chromosome that has TCAT repeated five times, and the chromosome from the other parent might have TCAT repeated seven times.

You might ask how this block repeat happens. Molecular biologists, believe it or not, say that it is not clear how it happens. They have some ideas, though, but it is not clear. There are 10000 of STR’s scattered throughout the genome, but for DNA fingerprinting purposes, we use typically 13 of them.

These 13 repeated sequences are polymorphic. This means that there is more than one type of repeats. I might inherit five repeats of TCAT at a certain location, and seven from my other parent. If we were all the same for this repeat number, they wouldn’t do any good. These different numbers of repeats are what set us apart.

To analyze DNA this way, however, we need first to make a population survey. We need to know the frequencies of the alleles. For example, five TCAT are present 50% of the time in the population, and seven is present 50% also. If somebody comes across and has ten would be a totally different individual.

Supposing we are dealing with two of these 13 short tandem repeat chains, and that they have three alleles: A, B and C. Let’s say that the A allele is frequent in one person in a hundred. The B allele is one in five. The C allele is the more common, four in five. For the second short tandem repeat: A allele one in ten, B one in two, C two in five. A, B and C are just the number of repeats. A for example might be five, and B seven repeats.

So, let’s review:

Short Tandem Repeat 1:
A: one in a hundred
B: one in five
C: four in five

Short Tandem Repeat 2:

A: one in ten
B: one in two
C: two in five

Here is the key argument for doing DNA identification in this way. It comes from Mendel and probability. For a person to be carrying both the A and B alleles of STR number 1, the combined probability is the product of the two probabilities. The combined probability of A(1/100) and B(1/5) is:
The probability for STR number 2 A(one in ten) and B(one in two) is:

So, what is the probability of having both of them at once? Yes! It is the product of their probabilities:

We’re now getting a pretty low probability. One in ten thousand! This is the probability of carrying the four alleles: A and B of STR 1, A and B of STR 2. Think about it. We have 13 different systems for identifying people. If we’ve got these different alleles, the probability of two people having identical alleles is vanishingly low. It turns out that we’re all virtually unique in these sequences. That’s why DNA fingerprinting is so useful in identifying people.