All living things possess DNA. It is this which ensures we can be determined as distinctly human - it also distinguishes us from all other people because a person�s individual appearance is matched by genetic individuality. We all possess specific DNA sequences which act as genetic fingerprints. DNA is found in every cell of a human � it can be found in blood, hair, skin and semen, all of which are often left at the scene of the crime by the criminal, as a sort of genetic fingerprint. They are also left by the victim on the body of the perpetrator, through physical contact. A scientist will analyse these materials, and a DNA profile can be produced. The DNA profile of each person is unique to him, and it is this piece of information that is of use to the forensic scientist. No-one�s DNA is identical to anyone else�s (excluding the case of identical siblings), so if a sample matches the sample taken from the scene of the crime, there is a high probability that they originate from the same person.

Genetic fingerprints have been recognised as a potentially powerful tool within crime solving. This may appear to be a breakthrough in the solving of crime and it does have its uses - however the use of DNA profiling is not as straightforward as it seems. There are many problems with this type of evidence, and its value within the courtroom is often questionable.

DNA profiling is a relatively new process, which was only developed in 1985 by Professor Alec Jeffreys. After its discovery, it was quickly realized how useful it might be in the context of forensic evidence, and for several years in its early stages, it was considered as infallible. This infallibility was proven to be wrong in the well-known case of Castro, where it was mutually decided by the prosecution and the defence that the DNA evidence was not sufficiently reliable to support the theory that there was a match or a non-match. This conclusion was found, not because the DNA evidence was bad, but because the process in which it had been created was wrong. There had been several faults in the way in which the DNA had been tested and interpreted, and so it was no longer reliable.

There are two techniques of DNA profiling. The technique that Professor Jeffreys developed is multilocus probing, where a large number of bands are revealed in one test. Such a test can, in theory, produce an identification with a high degree of certainty. However, the technique that is more suitable for forensic work is single locus probing. This produces only one or two bands at any one test. However, the degree of identification certainty can be increased by using many separate single locus probes. One of the reasons it is still preferred for forensic work above multilocus probing is that whilst multilocus probing is more complex, and finds a greater variety of bands, this same complexity makes it less sensitive, and so more DNA is needed.

In order to carry out a DNA profile, a sufficient sample needs to have been collected. Any blood, semen, hair or skin found at the scene of a crime can provide this, as long as it is not seriously degraded and there is enough of it. Conditions such as intense heat might have an adverse effect to samples. Having collected such a sample from the scene of the crime, a sample of the accused must also be collected, so that the two profiles can be compared. However, the comparisons do not necessarily stop there. It is also useful to collect a sample from the victim, so that one can be sure that the questioned sample is not from the victim. Also samples may be taken from other suspects to exclude them or incriminate them; samples from other victims of similar crimes may be taken, to study the effect of the crime on the actual analysis. Performing a profile like this should show whether or not two samples match, thus linking the accused to the scene of the crime.

The use of a DNA profile does not only include this process however. To be able to understand its relevance, the information has to be interpreted. It has to be decided whether the matching bands are sufficient to proclaim a match, as is the case in normal fingerprint evidence. This decision varies considerably from case to case. After it has been decided that the profiles are a match, the probability of random profiles matching must be shown. This involves advanced statistics.

Many parts of the DNA structure do not differ from person to person - for example, the part of the DNA stating that an individual has two eyes, a nose and a mouth. These are generic not only to humans, but most other living creatures as well. There are also parts of the DNA structure generic only to humans, and in the same way, there are parts that are generic to different races. Therefore, two Mexican men will have far more similar DNA than a Mexican and an Englishman. This is very important when it comes to working out the probability of two DNA profiles matching. The statistics that are used should only relate to the probability of the sample matching a sample from the Mexican accused, as opposed to any other Mexican, and not the probability of the sample matching a sample from the Mexican accused, as opposed to any man at all. This greatly decreases the probability that the DNA profiles are a match.

DNA profiling encounters problems in the courtroom as well as in the laboratory. DNA profiling is a science that both scientists and lawyers argue about. Whilst a prosecution scientist may say one thing about the interpretation of a DNA profile, the defence scientist may say another. A reason for this could be shown by example. Fingerprint evidence was first introduced by Sir Francis Galton in 1892 but it was another 20 years before it was accepted by the judiciary as valid evidence. Further study was necessary before it was judged as good evidence. When genetic fingerprinting was discovered, it was quickly established in the judicial world as being the truth, and for a long time it was thought to be infallible. Experience has shown this not to be the case.

DNA profiling is a very complex affair, with many different technical stages. At any one of these stages problems may occur that could invalidate DNA profiling evidence. These may include the quality, quantity and purity of the DNA collected. If the DNA sample has been left in adverse conditions for even a short period of 24 hours, the sample may become degraded, and the evidence would no longer be able to be tested. Adverse conditions are generally moist, warm conditions. However, if the source of the sample is a blood or semen stain, it will last much longer, and may be reliable many weeks after they have been deposited. If the quality of the DNA is degraded in any way, throughout the testing, the bands will not appear concisely, but will be smeared. Incidentally, when DNA profiling evidence is submitted as evidence, the test to show that it is not degraded must also be admitted.

The quantity of the DNA must also be correct. It is very difficult to do a DNA profile if the sample is not very big. Also, if the sample of DNA is small it may be entirely used up by the prosecution, leaving the defence with no ability to carry out their own tests. The purity of the DNA could be a potential pitfall in the whole process. A problem that might occur here is contamination of the sample. For example, what is considered to be a pure sample, obtained from a bloodstain for example, might prove to be a mixture of the victim�s blood and the assailant�s blood. Impurities may also be picked up from bacterial DNA at the scene of the crime, or from the laboratory. If this occurs the sequence that is obtained from the sample will include the information gained from the bacterial DNA as well as the desired DNA.

These problems concern the sample - but the testing process itself also has many potential pitfalls. To be able to be tested, the DNA has to be extracted from the tissue sample. This is done by precipitating it with 70% ethanol, rinsing the recovered DNA in ethanol, and then drying it in a vacuum. If it is dried too much, it will not dissolve properly, which will result in partial digestion and an inaccurate estimation of the concentration of the DNA. The way in which the DNA is broken down to be read in the form of bands is by digestion by restriction enzymes. If only partial digestion takes place, the positioning of the bands on the gel is distorted creating a deceptive DNA profile. Such partial digestion may be prevented by careful preparation of all the equipment and reagents involved, and meticulous prevention of contamination.

At every stage of the DNA profiling process, there are problems that might be encountered. These can be avoided by ensuring that the laboratory personnel are experienced enough to do a DNA analysis, since this is not necessarily the case, even if they are qualified to do more general tests. Care must also be taken to ensure all tests are carried out accurately to prevent distorted results leading to false conviction.

There are however, many successful cases involving the use of DNA evidence. The murder of Candice Williams by Patrick Joseph Hassett took place in 1978 before the revolution of DNA knowledge was in existence. Without this knowledge they did not have enough evidence to arrest anyone, as much of the available evidence was circumstantial. The body of Candice was found a day after she had been murdered - she had been strangled and sexually assaulted. Hair samples and semen samples were collected from the scene of the crime, and compared with other samples, but no matches were found. Hassett was questioned, but was not under police suspicion.

In 1980 after two more rapes in the area, more evidence came to light concerning the involvement of Hassett in the murder. Hair and saliva samples were taken and compared with those previously collected. There were slight discrepancies between the samples, but Hassett was still the main suspect. However, there was no evidence to substantiate this claim and so he was released without charge. Hassett then spent a period of time in prison, during which the police could not collect further samples from him, due to the PACE Act, 1984 regarding powers of the police in prisons. In 1991, Hassett came out of prison, and allowed them to take hair samples only. These hair samples were analysed in conjunction with the original swabs. This was a difficult task due to low quality and sparse swabs, but using the multilocus probe technique, the probability that the attacker was not Hassett was one in 3,200. This is not really very high, and so single locus probing was tried - this produced a likelihood ratio of one in 12,000 that the attacker was not Hassett. With this information, the case was taken to court, and despite defence attempts to invalidate the DNA evidence, the prosecution were successful - the DNA evidence being a direct factor in this decision.

Continue to Chapter Three.

Back to Chapter One.

Back to dissertation front page.