10% or 5% match window in DNA profiling

Abstract

Though the FBI repeatedly claims that the width of the window of the matching criterion they use in DNA profiling is 5%, it is shown that the width is indeed 10%. This fact has been noticed by some people but seems unknown to many. However, no one seems to be aware of the fundamental problem it creates to the fixed bin method employed by the FBI and some other laboratories because about half of the bins have sizes less than 10%. In other words, the probability for a random match of the DNA fragments from the crime scene and from the suspect could be underestimated. The problem may have serious implications for commonly adopted legal and forensic practices. The potential seriousness of underestimating the match probability is illustrated using the Hong Kong Chinese database.