Who has ever analysed the data in national DNA databases?  Courtesy of programs like CSI and the news/media, we all know that DNA can be an extremely powerful tool in crime solution.  When a sample from a crime scene is compared with a sample from an individual it should be the easiest thing to be able to say whether or not they originated from one and the same person, shouldn’t it?

In forensic science (and many other areas of science), without a solid and reliable database, interpretation of results can be troublesome or even meaningless.  So who has checked all the databases that have been built up by Police and prosecution agencies over the years?  Who checks them to make sure that the data is correctly entered or that the statistical basis for the interpretation is still correct and appropriate?  As it turns out, it doesn’t appear that anyone is really doing it, particularly in relation to the USA.  An article in this week’s New Scientist demonstrates this: Unreliable evidence? Time to open up DNA databases. Most of the world’s DNA results (and I am talking about those relating to criminal casework) are interpreted using the results of relatively small studies undertaken during the early years of DNA forensic casework.

The opening two paragraphs of the article cover it quite nicely, I think: “When a defendant’s DNA appears to match DNA found at a crime scene, the probability that this is an unfortunate coincidence can be central to whether the suspect is found guilty. The assumptions used to calculate the likelihood of such a fluke – the “random match probability” – are now being questioned by a group of 41 scientists and lawyers based in the US and the UK. These assumptions have never been independently verified on a large sample of DNA profiles, says the group. What’s more, whether some RMPs are truly as vanishingly small as assumed has been called into question by recent insights into DNA databases in the US and Australia. I find that slightly unnerving because I know from professional experience that if a DNA result is presented in court, it’s unbelievably difficult to shake anyone’s faith in the result.

I’m not suggesting that there is anything significantly wrong with any of the DNA databases but they should be open to examination by independent scientists so that the reliance that is placed on DNA interpretations can be shown to be well-placed. It makes the results relevant and reliable, which is important for acceptance as evidence in court.

The article does a good job of explaining the problems associated with DNA databases, describing three possible reasons why some profiles on the reference databases match one another (if they are all from different people then they should all be different): 1. duplication of the same profile in the database; 2. assumptions about frequency of occurrence of allelles in populations (see article for definition); and 3. the large number of relatives that may be on a given database (people who are closely related share more common areas of DNA in their profiles than people who are unrelated, which is logical). Particularly in relation to the latter point, if only a partial profile is obtained from a crime scene (i.e. only part of the DNA profile ‘picture’ can be detected), the chances increase of the sample matching more than one person’s database profile. (N.B.  There are essentially two sets of information in DNA databases – DNA profiles obtained from crime scene samples (so initially we don’t know who is the source of them) and DNA profiles taken from actual known people, usually suspects (depending on the legal jurisdiction)).

Of course, the converse of this is that if a DNA profile in the database has been incorrectly entered in some way, the chances of getting a “hit” on a cold case are also affected (a “hit on a cold case” is where re-examination of evidence is undertaken some time after a crime was committed, usually years afterwards, and a newly-acquired crime scene DNA result is run through the database for the first time.  If the crime scene profile matches a person’s profile already on the database, you’ve got a “hit”).  If the data’s not correct in the database, a ‘hit’ might become a ‘miss’.

The big question is whether the FBI will allow a group of independent scientists to review the USA’s CODIS database – and the answer at the moment is no.  This to me is a fairly typical attitude of prosecution agencies for this kind of access.  I can understand their reluctance in a way but I would think it would be a better thing to have it reviewed now rather than wait for a case to demonstrate a monumental stuff-up and get beaten with the consequences later, particularly to have it reviewed by people who would sign confidentiality agreements and then look at the data as a whole rather than focussing on one case where there may well be a problem.  As the closing commentary says, “Without external scrutiny of the databases, doubts will remain….All of this can be resolved by letting scientists have access to the data to do what they need to do.“  We all have our dreams, and I hope theirs comes true.

[The New Scientist article is based on a Science article, December 2009, v.236, p.5960]