1. Archive

DNA expert says he made a mistake

A computer programing error made by a prosecution scientist calculating the genetic frequencies in mixed bloodstains made those calculations "consistently wrong," the scientist testified Monday.

The genetic frequencies in the bloodstains are key state's evidence against O.J. Simpson, accused in the murder of his ex-wife and her friend.

"It was worse than I realized," said Bruce Weir, a world-renowned population geneticist. Weir said he thought he had made an error in only one of his calculations. Instead, he said he found the error ran through most of his totals regarding the number of people in the population who might have contributed to bloodstains said to link Simpson to the killings.

"I was consistent," Weir told jurors, "but consistently wrong." He later added, during prosecution questioning, that the changed tallies did not yield results significantly different from the first batch.

The stains are called mixed bloodstains because they contain evidence of more than one person's blood.

Defense attorney Peter Neufeld led the witness through a series of corrections showing, according to Weir's new figures, that it was now about 2{ times more likely that someone other than Simpson could have contributed to stains found in Simpson's Bronco and on a glove found at his estate.

For instance, three mixed stains on the Bronco's console were originally calculated to indicate a 1-in-1,400 chance that any two people could be responsible for them. The corrected chance, Weir said, is 1-in-570. Likewise, the chance of the glove stain being someone other than Simpson fell from 1-in-3,900 to 1-in-1,600.