Arguably, the reputation of DNA evidence has been based on the sound science that underpins it.  Widely accepted molecular and population genetics techniques provided the scientific basis of DNA profiling.  Single source profiles from identifiable body fluids enabled reliable inferences to be made.  Mixtures could sometimes be interpreted, but sometimes not.  Now, we have been involved in trials where, even though the DNA analyst is unable to provide a reliable statistic, evidence from computerised statistical programmes, devised by statisticians with an unclear understanding of the analytical process, and not yet widely tested or validated, have been introduced as evidence.  This is a potentially extremely dangerous development that not only threatens the confidence of juries in accepted techniques, but circumvents the entire thrust of measures to ensure the reliability of evidence in criminal trials.

In 2009, the National Academy of Sciences of the US produced a substantial review of forensic science.  It stated,

"With the exception of nuclear DNA analysis, however, no forensic method has been rigorously shown to have the capacity to consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual or source. In terms of scientific basis, the analytically based disciplines generally hold a notable edge over disciplines based on expert interpretation."

It is this that has earned DNA profiling the accolade as the 'gold standard' for forensic practice.

One ongoing problem in forensic DNA profiling is how best to assess the evidence from mixtures of DNA.  The complexities of mixture interpretation can be illustrated quite simply:  if I have profile AB and you have profile CD, our mixed cells would have a profile ABCD. However, the same profile could be produced by two people with profiles AC and BD, or AD and BC, so from a simple mixture of two people's DNA, at least six 'suspect' profiles can be created.  When considered across a DNA profile, the number rises to many millions.

One way of dealing with mixtures is a technique known as the Likelihood Ratio.  Even though this is not universally accepted as the best way to deal with mixtures in criminal cases, it is nevertheless commonly used and accepted in the UK for single source and clear two-person mixtures (i.e. when there is no evidence of a third person – estimates are always of the minimum number of contributors).  There is no generally accepted method of dealing with complex mixtures.

We now see in trials in the UK the introduction of statistics produced using 'sophisticated' software where the DNA expert has been unable to provide a statistic because the profile is simply too complex.  That usually means that the DNA expert has concluded that there is an uncertainty as to the number of contributors and/or the presence or absence of some components.

The International Society of Forensic Genetics stated in 2006 that,

"The implementation of such an approach in routine casework, in particular when using a computer-based expert system for mixture interpretation, requires an extensive validation of the variable parameters such as Hb and Mx, as well as appropriate guidelines for all laboratory procedures. ...

We note that accrediting standards such as ISO17025 require traceability, which may be interpreted as excluding ''black boxes''. This is a consideration in using expert system computer programs"

At least one of the systems that we have seen in casework has not, in our opinion, been subject to the necessary validation.  Indeed, the method does not use all of the information from the profile that an analyst would normally use to include or exclude an individual.

Without an effective challenge using, if necessary, international experts and the broader scientific community, these programmes will enter the justice system by default.  Time and again, and most recently by the Law Commission, proposals have been made to introduce proper testing of any technique to be used in Courts (the onus being on the proponent of the technique).  The NAS report stated,

"In a number of forensic science disciplines, forensic science professionals have yet to establish either the validity of their approach or the accuracy of their conclusions, and the courts have been utterly ineffective in addressing this problem."

There is a clear danger that the same is about to happen with the use of statistics in cases involving complex DNA profiles.

In the meantime, only prepared barristers, properly funded experts, and an appropriate forum will ensure that any system that is to be relied upon to prosecute cases will have been subject to the proper extent of testing and validation.

How likely is that?