Future directions of forensic DNA databases.

Abstract

Forensic DNA databases are indispensable tools of the law enforcement system. The purpose of establishing forensic DNA databases was to develop investigative leads for solving crime and usually was the purview of “criminal justice agencies for law enforcement identification purposes” (1). The forensic DNA databases of most countries generally contain two types of profiles: 1) reference profiles from convicted offenders and/or arrestee profiles; these profiles are from known sources; and 2) forensic profiles which derive from crime scenes and are characteristically from unknown sources. In a typical database search, an unknown forensic profile is searched against the convicted offender and arrestee profiles (or can be searched against other unknown forensic profiles) to determine if an association, often called a match or hit, can be found. The hit can be used to develop investigation leads. As of May 2013, China and the United States (US) maintain the two largest forensic DNA databases, containing more than 20 and 12 million profiles and have produced over 410 000 (2) and 185 000 hits (3), respectively. In addition to direct matching between known and unknown sample profiles, profiles from missing persons and their relatives, as well as unidentified human remains, are included in a number of databases (2,3). Missing person identification also is an invaluable module for investigating certain crimes. For example, as of June 2013, China has successfully identified and rescued 2455 trafficked children through the use of its DNA database (2). The US National Missing and Unidentified Persons System (NamUs), which uses other meta-data has successfully solved 3499 missing persons cases as of August 2012 (4). As expected with the great success of the use of forensic DNA databases, new challenges are emerging. The databases are experiencing rapid growth, and thus there is a potential of increased adventitious hits; the power for current and new applications (eg, missing person identification and familial searching) require additional infrastructure support; and there is an increased desire for international data sharing (5-7), which possibly could be hampered if only a relatively small number of loci is shared among laboratories worldwide.