YOU ARE HERE: LAT HomeCollections

Blood Amount for Tests Is Crucial : Simpson case: Most of the specimens on which the prosecution is building its case barely meet minimum requirements for accurate DNA results. A second test requires a smaller sample, but the results are far less definitive.


Drops of blood smaller than a grain of rice may be the ultimate arbiters of guilt or innocence in the O.J. Simpson murder case, experts said Monday.

And as pretrial arguments swirled around the use of genetic blood testing, it was clear that the amount of blood available for such tests will be crucial, according to scientists in the field.

One small drop, which would be about the size of a dime if smeared onto clothing, is the minimum amount necessary to obtain viable genetic fingerprinting using the most accurate DNA test, called RFLP.

Prosecutor Marcia Clark is relying on the esoteric technology to try to prove that blood drops on a sidewalk at the townhouse where Simpson's ex-wife and her friend were killed belong to Simpson, and that bloodstains in his Ford Bronco and on a glove found at his Brentwood estate belong to the victims.

Simpson's defense team wants to obtain a portion of each blood sample to conduct its own testing, but most of the blood specimens on which Clark is building her case barely meet the minimum size requirement. Some are smaller.

It might be possible to obtain results with the smaller samples, according to geneticist George F. Sensabaugh Jr. of UC Berkeley, but the chances are poor that the results would be useful. In Monday's hearing, Clark said it is unlikely there will be enough left over for independent testing by the defense.

For those blood specimens not large enough for RFLP fingerprinting, the prosecution will fall back on a second, less specific type of fingerprinting called PCR. It requires a far smaller sample and is performed much more quickly, but its results are not nearly as definitive. Clark may also find it harder to have PCR results admitted into court because the technique is newer than RFLP and has not been used in court nearly as often.

RFLP, an acronym for Restriction Fragment Length Polymorphism testing, is a well-studied, time-consuming test that can show with a very high degree of certainty--at least 999,999 chances in 1 million--that two samples are from the same person. The test, first developed in 1985, looks at four to six segments of DNA that have a unique length in each individual.

RFLP requires 50 nanograms of DNA. That is not very much blood, according to Dr. Paul R. Billings of the California Pacific Medical Center in San Francisco. One milliliter of blood, an amount about the size of the eraser on a pencil, contains 20,000 nanograms.

PCR testing requires a stain only about the size of the head of a pin, according to Mark Stolorow of Cellmark Ltd. in Gaithersburg, Md., which is performing DNA testing for the LAPD in the Simpson case. In this technique, a biochemical process called Polymerase Chain Reaction is used to replicate segments of DNA, increasing the quantity a million times or more.

But PCR does not provide as much information as RFLP. The most commonly used form involves a series of six markers called, collectively, DQ-alpha. PCR tests using DQ-alpha have been admitted in court in more than 80 court cases in 23 states, according to Kristin Garvin of Perkin-Elmer Corp., the test's manufacturer. But even under the best of circumstances, DQ-alpha provides odds of no better than 93 in 100 that two samples are from the same individual.

PCR is better for showing somebody did not commit a crime than for showing guilt, according to criminalist Jack Mertens of the FBI's Forensic Laboratory.

Reported PCR results for the glove found on Simpson's estate, for example, show that the blood on it could have come from either Simpson or his ex-wife. But the results were so poor that the blood could have come from 50% of the entire population, said forensics expert William Thompson of UC Irvine.

Presumably because of this, Clark said that the prosecution wants to do a different type of PCR testing on the same samples, using a set of probes called "polymarkers." These probes are more definitive than DQ-alpha, but are much newer and have not been accepted in as many court cases, Sensabaugh said.

Given the high reliability of RFLP testing, perhaps the best hope of the defense lies in finding errors in Cellmark's testing. Shapiro's co-counsel, Barry Scheck, said Monday that the company had made a mistake in a 1989 proficiency test--in which Cellmark was asked to match standardized blood samples whose identities were unknown to its scientists.

In an interview with The Times, Stolorow conceded that error, but noted that the company had tightened its procedures substantially since then. "We are now involved in every available externally offered proficiency test for forensic DNA analysis, and we have made no errors in the last 300 to 400 test samples over the last five years," he said.

Sleuthing With DNA

Los Angeles Times Articles