DNA markers in Apartment 5a

Maybe this is already posted somewhere, but since mention of Eddie & Keela has been in The Telegraph today and reference to the DNA markers, thought I'd pop this here for reference:


Re: 15 out of 19 markers
Post AnnaEsse on Mon 12 Mar - 9:44

I am going to copy part of an article, which appeared in a blog, which has disappeared now because of the owner's having 'family issues.' Enfants Kidnappés blog featured details and news on many missing children and it is a loss to the online community. I have quite a few articles from there on my blog and I am grateful for the work put in by the owner in keeping it going for so long.

The article I am quoting from was published in August 2008 and features an interview with Madame S. Adamis, Legal Expert at the Centre for Human Genetics at the Catholic University of Louvain. (GNEX - UCL)

The DNA results.

The media war has begun. The British newspapers do not speak the same language as the Portuguese newspapers. Thus, Clarence Mitchell, the parents' spokesperson, states in the British press:

"The DNA results were never 100% compatible with Madeleine's DNA. A note of caution had been expressed at the outset. The police were wrong to pursue this line of investigation. The Portuguese judicial system has admitted having no evidence! I can confirm that the PJ stated to Gerry that Madeleine's DNA had been found in the two apartments and in the vehicle, while it is now clear that that is not what the first report from FSS said. You must ask yourself what the police were trying to do in inventing evidence which they don't have and which they cannot have. In these circumstances, it could be asked what is their motivation."

We are going to flatten Mr Mitchell's statements somewhat. From the point of view of a police officer, it happens that the police tell suspects that they hold evidence, which they don't in reality have, in order to crack a suspect; this should not be done any old way. Indeed, it is difficult to tell a suspect, for example, that his fingerprints were found at a crime scene while the suspect knows that he appropriately wore goves throughout his offence and that as a result the police are talking rubbish! You must be quite serious. Putting forward evidence which you don't have must remain plausible, knowing that if the suspect is indeed the guilty party, the police will lose face if they invent evidence that the suspect knows to be false!!

On the other hand, Clarence Mitchell's interpretation of the DNA results invite circumspection. To help us with our thinking, our association called upon an expert on the subject. She is Madame S. Adamis, Legal Expert at the Centre for Human Genetics at the Catholic University of Louvain. (GNEX - UCL)

Association Enfants Kidnappés: Do you know about the LCN technique and is it commonly used?

S.Adamis: LCN (Low Copy Number) is a technique developed by the laboratories of the English Forensic Science Services to analyse samples containing a very small number of molecules of DNA. The basic principle is to increase the number of PCR cycles to increase sensitivity. [Gill, 2000; Whitaker, 2001] This technique has two major drawbacks:

1) It produces unbalanced profiles for one or more markers, with possible disappearance of an allele due to the stochastic* effect.

(*Involving chance or probability)

2) It leads to the detection of one or more alleles in negative controls of unknown origin.

The first drawback leads to obtaining an incomplete profile, partially wrong and not reproducible. This low reliability goes against principles of quality advocated notably by standard ISO 17025 in force in our country.
The second requires necessarily working in special conditions to avoid contamination inherent in the environment and particularly from human DNA present in the dust in the atmosphere or on the surfaces of objects. Given the limitations of this technique, the conclusions produced could easily be attacked or invalidated during a trial. This technique is not routinely used in laboratories for genetic identification in Belgium.

AEK: If in a DNA analysis, 15 out of 19 markers belong to person, "x", can we conclude that it is indeed from that person?
S.A.: If the profile is complete and of quality, and the analysed markers are informative, then without doubt! The result is discriminating. This result is very very reliable. The error is in the region of 1 in a billion! It is almost impossible for it to be otherwise. For a convincing DNA profile, there must be a minimum of 7 base markers. In the case which you present, 15 markers out of 19 does not leave any hanging doubt. This result is completely reliable and usable in court. The error rate of one in a billion is so unlikely that the results are accepted by magistrates without lawyers being able to place them in doubt.

We are going to digress here. We understand even better the attitude of the PJ. Indeed, faced with the inconsistencies we are talking about above, the police officers have doubts. Then the dogs detect traces of blood and cadaver odour. The doubts transform into beliefs. But something is missing. All these elements are not sufficient to face charges in a court. Confirmation is lacking. Scientific confirmation. That confirmation arrives with the first report of analyses which states that 15 out of 19 DNA markers belong to Madeleine. That's all the police need. The evidence is there. Obvious. The first report of the analyses proves the parents' guilt in the eyes of the investigators. This report would be considered as irrefutable proof by, I believe, all the police. From then on the parents were placed, logically, under the specific status of, "arguidos." Of course, an error rate of 1 in a billion is not a 100% profile, in that Clarence Mitchell is right. Then, afterwards, comes a thunderbolt.* A second report from FSS arrives and totally contradicts the first. Also ruining the evidence the police thought they had. According to this report, the harvested samples would have been contaminated, making them very unreliable in the end. Several DNAs would have been mixed, creating the DNA of anyone!

* I took liberties there with the interpretation because that was the best phrase that came to me in the context!


EAK: If we harvest the DNA of three different people, can we recreate the DNA of anyone?

S.A.: If the harvested DNA is mixed with the DNA of three individuals in a balanced way, then effectively we could find the DNA profile of anyone. Mine just like yours. But the mixture must be perfectly balanced, otherwise the rate of reliability becomes very low, going from 1 in a billion to 1 in a thousand or 1 in a 100. It is no longer a question of considering this result as discriminating.

Ok, what happened? LCN unreliable? The samples contaminated during the second analysis but not in the first? The first analysis botched by FSS? There would be a good way of finding out. When two experts contradict each other, a third analysis is carried out in an independent lab. But, alas, following an incident that remains unexplained, the only existing DNA samples were, unforunately, lost or destroyed by the laboratory, which renders impossible a third analysis!

That's a really informative article Bluebell, thank you! What I would like to know is does the reference sample degrade each time you test it, making each subsequent test less reliable?