Interpreting DNA Evidence: Can Probability Theory Help?

The interpretation of DNA evidence has been the subject of substantial controversy, both in the courtroom and in the scientific and legal literatures. The debate revolves around a number of topics, including population genetics issues, the role of possible laboratory errors, the effect of database “trawls” on evidential weight, and criteria for establishing the “uniqueness” of a profile. In large measure, the confusion surrounding the controversy can be attributed to the concepts of classical statistical inference, with which most scientists are familiar, but which are inappropriate in this setting. I argue for an affirmative answer to the question in the title, showing how many of the areas of controversy are readily resolved by insights obtained from probability-based reasoning focused on the relevant question, which is typically, Is the defendant the source of the crime scene DNA? The argument developed here is inspired by my experiences of advising courts, lawyers, and defendants in many UK criminal cases, in addition to researching relevant statistics and population genetics issues.

[1]  S Sawyer,et al.  DNA fingerprinting loci do show population differences: comments on Budowle et al. , 1996, American journal of human genetics.

[2]  Eric S. Lander,et al.  DNA fingerprinting dispute laid to rest , 1994, Nature.

[3]  N. Kerr.  Stochastic models of juror decision making. , 1993 .

[4]  P Donnelly,et al.  DNA database searches and the legal consumption of scientific evidence. , 1997, Michigan law review.

[5]  D J Balding,et al.  When can a DNA profile be regarded as unique? , 1999, Science & justice : journal of the Forensic Science Society.

[6]  D. Petrov,et al.  Genetic differences at four DNA typing loci in Finnish, Italian, and mixed Caucasian populations. , 1992, Proceedings of the National Academy of Sciences of the United States of America.

[7]  Ian W. Evett,et al.  The impact of recent judgements on the presentation of DNA evidence , 1998 .

[8]  D. Balding,et al.  Significant genetic correlations among Caucasians at forensic DNA loci , 1997, Heredity.

[9]  Eric S. Lander,et al.  DNA fingerprinting on trial , 1989, Nature.

[10]  Christopher Anderson.  FBI attaches strings to its DNA database , 1992, Nature.

[11]  B S Weir,et al.  The effects of inbreeding on forensic calculations. , 1994, Annual review of genetics.

[12]  D. Hartl,et al.  Population genetics in forensic DNA typing. , 1991, Science.

[13]  Dj Balding.  Errors and misunderstandings in the second NRC report , 1997 .

[14]  D. Balding,et al.  Population genetics of STR loci in Caucasians , 2006, International Journal of Legal Medicine.

[15]  Ian W. Evett,et al.  Bayesian Analysis of DNA Profiling Data in Forensic Identification Applications , 1997 .

[16]  David J Balding,et al.  Effects of population structure on DNA fingerprint analysis in forensic science , 1991, Heredity.

[17]  I. Evett,et al.  Interpreting DNA Evidence: Statistical Genetics for Forensic Scientists , 1998 .

[18]  J. Koehler.  Error and Exaggeration in the Presentation of DNA Evidence at Trial , 1993 .

[19]  B. S. Weir,et al.  A bibliography for the use of DNA in human identification , 1995 .

[20]  Dj Balding.  Forensic applications of microsatellite markers , 1999 .

[21]  D J Balding,et al.  Inferring identify from DNA profile evidence. , 1995, Proceedings of the National Academy of Sciences of the United States of America.

[22]  D J Balding,et al.  Evaluating DNA profile evidence when the suspect is identified through a database search. , 1996, Journal of forensic sciences.

[23]  Ian W. Evett,et al.  Avoiding the transposed conditional , 1995 .

[24]  K. Roeder.  DNA Fingerprinting: A Review of the Controversy , 1994 .