Weir B S
Department of Statistics, North Carolina State University, Raleigh 27695-8203.
Proc Natl Acad Sci U S A. 1992 Dec 15;89(24):11654-9. doi: 10.1073/pnas.89.24.11654.
The use of matching variable number of tandem repeat (VNTR) profiles to link suspects with crimes is potentially very powerful, but it has been quite controversial. Initial debate over laboratory procedures has largely given way to debate over the statistical and population genetic issues involved in calculating the frequency of a profile for a random member of a population. This frequency is used to weight the evidence of a match between suspect and crime scene material when the suspect denies responsibility for that material. A recent report from the National Research Council, intended to put to rest some of the issues, has instead raised further debate by advocating a procedure based on maximum frequencies of profile components over several different populations.
利用匹配的可变数目串联重复序列(VNTR)图谱将犯罪嫌疑人与犯罪联系起来,这一做法可能非常有效,但一直颇具争议。最初关于实验室程序的争论,很大程度上已让位于关于计算群体中随机个体图谱频率所涉及的统计和群体遗传学问题的争论。当犯罪嫌疑人否认对某物证负有责任时,该频率用于权衡犯罪嫌疑人与犯罪现场物证之间匹配的证据。美国国家研究委员会最近的一份报告旨在平息一些问题,然而,该报告倡导基于几个不同群体中图谱成分的最大频率的程序,反而引发了进一步的争论。