Discrimination and DNA databases

Last month, the South Australian government began accessing the national DNA database. In just the first few days the database has linked suspects to more than 20 unsolved crimes, including rape and assault, some dating back to the 1980s. Currently, seven of the eight Australian states and territories have agreed to share the information on their DNA databases using the National Criminal Investigation DNA Database run by CrimTrac.

The national database does not contain the whole DNA sequences for each person but rather a set of characteristics which can be used to establish whether two samples came from the same person.

Only 2% of the human genome is used to determine a person's biochemical make-up and physical characteristics. In some sections of the non-coding parts of the genome there are repeated patterns of nucleotides with the number of repeats in each sequence varying from person to person.

For forensic typing, scientists consider the number of repeats at several positions, or loci, on the genome. As each locus has two alleles there are two numbers generated for each locus. These sets of numbers, combined with the gender, create the DNA profile for each person.

In Australia, the DNA database is based on nine loci of short tandem repeats (STR) which are four to six nucleotides long; in the UK, the Home Office currently relies on STRs at 10 loci and in the US the Combined DNA Index System is based on STRs at 13 loci.

The DNA profiles are proving an effective means of identifying and clearing suspects but obviously enough, success depends on the contents of the database. For example, in the UK, 37% of black males, 13% of Asian men and 9% of white men have had their DNA profiles included in the national database. These figures portray an even bleaker picture when the ages of the men included in the database are taken into consideration. Then about 77% of black males aged between 15 and 34 are in the national database while only 22% of white males in the same age bracket are included.

New trends in database searching are exacerbating the discrimination issue. It has been realised that a 'near-match' may be a close relative of the crime suspect. Familial matching involves searching the DNA database to determine if a crime scene profile can be linked to a blood relative of a person on the database. These familial searches have already yielded results in the UK, US and Canada.

In Australia, CrimTrac is currently examining the numerous legislative, ethical, privacy and technical issues surrounding familial matching.

Familial DNA profile matching can be used in some positive ways; for example, in victim identification. Kinship matching is the use of DNA profiles to link blood relatives in order to assist in civil investigations. In exactly the same way as familial matching, blood relatives' profiles can be linked because half a person's nuclear DNA is inherited from their mother and half from their father. Kinship matching is extremely valuable for assisting with missing persons and unknown deceased investigations, and disaster victim identification.

One way to overcome bias and discrimination in the DNA database content is to include the entire population in the database. In 2005, Portugal was the first country in the world to announce its intention of including its entire population in its database of DNA profiles.

If this happens here, the entire concept of 'innocent until proven guilty' will be overturned and issues such as DNA planting at crime scenes and false-positive errors from the forensic laboratories will have the potential to destroy lives and reputations.