Unresolved issues in the forensic use of DNA profiling

The enormous amount of genetic diversity in humans allows for a powerful form of individual identification. This “DNA profiling”; is based on the fact that sites within the human genome have variable numbers of tandem repeats (VNTRs) and has been hailed in forensic sciences as the greatest discovery since fingerprinting. The techniques involved are virtually the same as those used in all molecular biology laboratories. A major difference however is that in forensic science DNA samples can be less than ideal in both quality and quantity. Furthermore, in basic molecular biology the origin of the sample is known while in forensic testing it is not. Thus, the challenge is to reconcile a “match”; between a crime scene DNA sample and one from a suspect(s). Presently, a debate exists regarding the use of the unmodified product rule versus a more conservative ceiling principle approach to calculate the probability of a coincidentally matching DNA profiles. The latter was endorsed in a recently published report by the prestigious National Research Council but has not received widespread support from testing laboratories. Further exacerbating the debate over how much weight should be attached to DNA profile evidence is a lack of widely accepted standards for forensic laboratories especially in the areas of proficiency testing, publication of error rates and laboratory personnel certification.