The O.J. Simpson trials were perhaps the lengthiest and most detailed trials to ever hit the American judicial system. This was partly true to the multitude of expert witnesses, one of whom was a professor here at N.C. State. Bruce Weir, a professor of Statistical Genetics, is one of the foremost experts in the country on how to interpret DNA evidence in the courtroom. The purpose of his testimony was to provide numerical weight to the evidence of matching DNA evidence.
DNA is the genetic material contained in every human cell. In the past decade extracting DNA from blood stains, hair, bones or semen has become an increasingly common way of identifying supects in a crime. However DNA identification can not yet be presented in courts as either a unique positive or negative match. Like fingerprinting in its first 30 years, DNA data must come with statistics that give the chances of finding a random person with a matching DNA profile to the one found on the crime scene. In other words, if you find a blood stain and extract a DNA sample, what are the odds of finding another DNA sample identical to that one from a completely different person?
Weir's courtroom work involves calculating these probabilities which often come to extremely large numbers. In the Simpson trial, the DNA profile from a blood stain on the rear gate was expected to occur only once in 57 billion people. Since the world population is only 5 billion, it is basically impossible that someone else could turn up matching this profile given the fact that someone(O.J.) already matches it.
The Simpson case was groundbreaking in the fact that it involved interpreting multiple DNA profiles since some of the blood stains on the evidence apparently came from several different people mixed together. The technique for calculating statistics for mixed stains is distinct from that of calculating stains from only one source and Weir remains one of the few statisticians in the country able to do this complex calculation. In fact this method of calculation was not yet endorsed by the National Research Council at the time of the original Simpson trial, a fact which the defense team brought up to question his credibility . It was accepted in the updated report, published in April 1996, where Weir is the most cited scientist in the publication.
Weir is quick to point out that his numbers are not the probabilities of guilt and innocence, as is sometimes misconstrued. His numbers tell nothing about the possibility of errors in collection or handling of the samples, something he says is impossible to give. The defense team in the Simpson trial took advantage of this to cast doubt on his calculations because of allegations of police misconduct in handling the evidence.
Another misunderstanding that often arises both in the courtroom and in the media is what Weir calls "transposing the conditional" or the "prosecutor's fallacy". This subtle flip in logic greatly changes the meaning of the data.
For instance, in a blood stain found on the Bronco, the DNA profile showed that there was a 1 in 1400 chance that the profile could have come from someone other than O.J. Simpson or Ron Goldman.
But an Associated Press report misinterpreted this statement to say that there was a 1 in 1400 chance that ANY two people in the population could be responsible for such a profile. Weir compares this error to saying "All four legged animals are dogs" instead of "All dogs are four legged animals".
Coincidentally , William Thompson, one of the defense attorneys to question Weir, was the person that coined the phrase "prosecutor's fallacy" and tried to use it to confuse Weir on the stand although normally Thompson usually strongly cautions against misinterpreting the data in this fashion.
Another item that the defense team used was a mistake Weir used in his calculations which he quickly corrected. He says that the mistake was relatively minor and attributable to the fact that the court demanded him to do additional calculations which he did in his Los Angeles hotel room and did not have time to check over before he presented them. On the stand during cross-examination, he showed regret, saying "I'm going to have to live with that mistake for a long time".
This mistake apparently hasn't hurt Weir's career or reputation very much. He was invited to speak on a panel this month at the American Association for the Advancement of Science in Seattle on his research on the use of DNA in the courtroom. His expertise continues to be in demand for trials all over the country. A US Air airplane ticket sits on his desk in silent testimony to his trip to Philadelphia last week where his results helped get an accused rapist convicted.
Weir refuses to accept payment for his court testimonies except for travel expenses. When asked why he testified, he answered "I'm very concerned that the statistical interpretation of DNA evidence is done correctly. I've been very angry over the last four years at some of the statements made in courts about the statistics attached to DNA profiles. I am very anxious that they be done correctly and I am therefore willing to testify when possible". Eventually, Weir hopes his job in testifying will become obsolete. He points out that the numbers he comes up with are often in the millions or billions which are hard to understand. In a recent commentary published in "Nature Genetics", he concluded "I look forward to the time when DNA profiles are of such extent and integrity that they are recognized as being as probative as fingerprints". The 1996 National Research Council report backed him up on this saying "We foresee a time when each person(except for identical twins) can be identified uniquely."