Two of the most important sorts of molecules in keeping you, and just about everything else on Earth alive are DNA and proteins. There are other molecules and substances that are important, but these two are the ones that you've probably heard of, and we aren't here to teach you all about biochemistry, we will leave that to the specialists.
DNA is a long
chain of a molecule that roughly resembles a piece of string. Actually,
it more closely resembles two pieces of string joined together all along
their length, and then twisted in a 'Double Helix' (a helix is a spiral,
a bit like a helter-skelter), and DNA is called a double helix because
the two pieces of 'string' wind round each other in this helix shape.
The 'bits' of DNA that go to make up the string are called nucleotides.
Each nucleotide contains a piece (sugar-phosphate) that goes to make up
the 'backbone' of the string, and a piece (called the base) that
stretches across between the string and joins to a similar piece
stretching out from the other strand of the DNA molecule. There are four
types of bases called Adenosine (A), Cytosine (C), Guanine (G) and
Tyrosine (T). These bases only join to a specific other base, so C and G
will join together, and A and T will join together, but C won't join to
C, A or T and so forth. You can see this in the picture, if A is Yellow,
it only joins to T (red), and similarly C (Green) only joins to G
(blue).
Almost all of the cells in your body have DNA in them,
divided up into 46 lengths, called chromosomes. Everyone has 22 pairs of
chromosomes (making 44 altogether) and 2 chromosomes that determine your
sex. This could be two X chromosomes (as in the picture), or an X and a
Y chromosome. If you have two X chromosomes you are female, and an X and
a Y makes you male.
If you were to take the DNA out of a single cell and put it all
together, end-to-end, it would make a very thin string two metres long.
Because you've got DNA in just about all of your cells, if you took all
the DNA in your body and joined it together to make a really thin rope,
it would stretch from here to the moon and back 800 times!
Something that we hear about in the news from time to time is genetic fingerprinting. Genetic fingerprinting involves looking at someone's DNA to compare it to other DNA. Because parts of our DNA are unique (unless you've got an identical twin), this might be to prove whether or not that person was at a crime scene, or to make sure that someone really is the father of the person in question. The same sorts of techniques are also used in genetic screening which is a test that doctors can use to see if you are at risk of certain types of disease.
Genes are bits of DNA that are the code that we use to make proteins. Although it is the proteins that do the work, our DNA controls what proteins we have, and how much of them, which is why people have different colours of hair, eyes and so on. There is a lot more variation in the DNA than you might think, partly because there is lots of 'junk DNA' that we don't use, and so any changes there are unlikely to affect our chances of growing up and passing it on to our children.
There are two main methods of genetically fingerprinting someone, one is called Restriction Fragment Length Polymorphism (RFLP) and the other is Microsatellite Polymerase Chain Reaction (mPCR).
Restriction Fragment Length Polymorphism (I will call it RFLP from now on, it is much easier to say and read) is based on the knowledge that bacteria make restriction enzymes, which are special proteins that cut DNA up in specific places. The bacteria do this because they can swap DNA between themselves and other sorts of bacteria, but they need to be able to control what they receive like this, and so they chop up most of the DNA unless it comes from something like them. But the enzymes aren't too fussy, and we have learned how to take them out of the bacteria and use them in the laboratory, and they cut up any sort of DNA quite readily.
So once you get your DNA sample you have to take it out of the
cell, and chop it up with these enzymes, which takes a bit longer to do
than it does to read, but not that much longer. Restriction enzymes look
for 4, 6 or 8 base sequences, for example there is an enzyme called
BamH1 which only chops when it finds a sequence GGATCC. If you look at
an enzyme which chops up DNA every time it finds the right 6 bases
together, then you can predict that it will chop up the DNA, on average
every 4 X 4 X 4 X 4 X 4 X 4 = 4096 bases. (Work out for yourself how
often an enzyme that chops every time it finds the right four or the
right 8 bases together would chop.) That sounds like quite a lot, until
you find out that human DNA has about 6 billion bases in it, and so this
would give you about 1.5 million fragments to look at! (The picture to
the right might give you some idea how confusing that could be.) The
good thing is, that because of all the junk DNA, and random mutation,
some of these fragments are different lengths in different people,
whilst some are probably close to being the same length, as the cuts
will occur in important genes where there isn't any variation. These
different lengths are why this technique is called by its name. The
Restriction enzymes make Fragments of DNA of different Length.
Polymorphism is a jargon word that scientists use to mean 'having many
shapes' or just 'different'.
If you've ever seen a genetic assay done
you will know that you don't get 1.5 million bands on the specimen that
you see. This is because it would be so hard to compare them all, that
it would be useless. Instead what the scientists do is spread out the
DNA fragments according to their length, on a gel. They can do this
because DNA has negative electrical charge, and so if you put it in a
current it will tend to move. The type of gel that is most commonly used
is made from an extract of seaweed, and it really does resemble jelly,
although it is usually an unappealing grey colour. This gel slows down
big fragments of DNA more than little fragments, and so the fragments
separate according to their length. But you've still got 1.5 million
different fragments, and so the final step is to stick a marker to some
of the fragments, but not all of them. This is done by a special process
called the southern blot (named after a Dr Southern, not the direction),
which sticks the DNA onto a nylon film, and then splits up the double
helix. Once you've got single-stranded DNA like this, you can add short
lengths of single stranded DNA to it, with a marker (a radioactive or
fluorescent tag on the end) and because DNA likes being in double
strands every time it finds a match the marker will bind every time it
finds its match. If you use a probe with about 13 bases you get about 60
bands, and if you use one with 15 bases you get about 15 bands, which is
much easier to see similarities and differences with. The picture to the
left shows a typical result of a southern blot with a radioactive tag.
It might not be as pretty as the picture above, but it is a lot easier
for people to understand.
You can do this whole process with either different enzymes, or different probes, or both, to make sure that you really do cover all the bases, and stand a good chance of answering the initial question.
Microsatellite Polymerase Chain Reaction (mPCR) uses a slightly different approach, but the same sort of idea. Microsatellites are regions of junk DNA which are filled with lengths of two base repeats, for example
GTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGT
The number of GT repeats varies between individuals, but since you inherit them from your parents, is likely to show a clear pattern within a family. PCR is a method that allows you to take a small amount of DNA and amplify it (make more of it, or certain bits of it). PCR works by repetitively melting the DNA (if you warm DNA up the joins between the strands loosen up and the strands come apart, a bit like melting ice), and then lowering the temperature slowly to allow a short length of DNA called a primer to bind to a specific piece of the whole molecule (just like ice will reform if you drop the temperature once more. Once the primer is attached there is an enzyme that will lengthen the DNA adjacent to the primer that is used to make the DNA stretch out. This lengthening process is perfectly normal, in fact your cells do it every time they divide, and cell division is part of the process that helps us grow, and stay alive. Actually for this purpose you use two primers, one for either end of the sequence that you are looking for, so that you selectively amplify up the interesting piece of the DNA.
As you cycle through this process several times you generate a small amount of 'noise' (DNA fragments that are a bit longer or shorter than most) and a very large amount of 'signal' DNA of the same length. In fact each cycle makes two pieces of noise, but doubles the amount of signal, so after 30 cycles (which is a typical overnight PCR experiment) you would generate 60 noise lengths, but about 1 billion bits of signal DNA, giving you a clear response. You still need to see how long the DNA fragment is, so you can compare it to the other pieces of DNA to see if they are the same.
If you read the description of RFLP above, this part might seem familar, but just in case: What the scientists do is spread out the DNA fragments according to their length, on a gel. They can do this because DNA has negative electrical charge, and so if you put it in a current it will tend to move. The type of gel that is most commonly used is made from an extract of seaweed, and it really does resemble jelly, although it is usually an unappealing grey colour. This gel slows down big fragments of DNA more than little fragments, and so the fragments separate according to their length. You can then look at the fragments, usually by staining them, or by having a dye molecule attached to the ends of the primer, and can easily compare the lengths.
These methods are quite useful, but not without pitfalls. However in some cases, especially paternity cases where the mother and presumed father also provide a sample 3 RFLP probes or 6 PCR primer sets are thought to give a 99% indication of paternity, and 6 RFLP probes or 9 PCR primers give greater than 99.99% indication. Information in a criminal trial about probability that the person is the only one who could have left the original sample is actually a lot more problematical, it relies on assumptions that have not been tested about how different patterns of inheritance of these markers occur in the different races, but it can still be a powerful tool to assist in crime fighting.
Click here to go back to the index.