Back to Publications

Failing to interpret DNA mixture evidence

Perlin, M.W. Failing to interpret DNA mixture evidence, Cybergenetics White Paper, July, 2016.




Mixtures are a powerful form of forensic DNA evidence. When two or more people contribute their DNA to a biological item, science can statistically establish that they were present on the same object. With rape kits, such co-location can place an assailant’s DNA on a victim’s body. Handgun mixtures of many people can show who did or didn’t handle a weapon. A shared hat dropped outside a bank can identify the robbers.

Crime laboratories generate informative mixture data, but do not always interpret them properly. Misinterpreting DNA evidence produces inaccurate match statistics, or gives no statistic at all. Unfounded statistical methods have failed on hundreds of thousands of mixtures.

The result is injustice for defendants denied potentially exculpatory evidence, injustice for victims whose cases are lost when inculpatory evidence is unreported, and injustice for innocents victimized by crime that DNA could have prevented.

This article reviews the history of DNA mixture interpretation failure. We begin in 1985, at the start of the genomics revolution, discussing the origins of modern DNA testing. Proceeding in five year increments, we outline the missed opportunities and policy failures that have brought us to the current situation. We offer recommendations to help overcome long-standing DNA interpretation problems.