Blood Work: The Man Behind the Crime Scenes

Dr. Lawrence Kobilinsky paves the way for a more exact forensic science.

Almost overnight, forensic science is becoming part of American popular culture. Turn on a television, and you can't avoid it: Court TV panelists discuss blood evidence, trace evidence, and DNA testing. On a series known as Body of Evidence, a retired criminal profiler details her most puzzling real-life cases. The lead character in one of several CSI programs scours a Manhattan dog show for clues to a woman's mysterious death.

The truth behind the Hollywood version of crime-solving, however, is that actual forensic detective work is rarely as exciting, speedy, or infallible as it's usually portrayed to be in popular entertainment.

"It's not as you see it on television," chuckles Dr. Lawrence Kobilinsky, Professor of Criminal Justice and Biochemistry at The Graduate Center and John Jay College of Criminal Justice, and Senior Science Advisor to the President at John Jay. "It's very good entertainment, and I enjoy watching it, but they make some significant mistakes. They do it because they need to make crime solving seem more glamorous."

It's Kobilinsky who serves as John Jay's point man on a range of scientific issues relating to criminal justice, as he shepherds everything from a more rigorous method of bite-mark pattern comparisons to a better way of doing the excruciatingly complicated lab work involved in matching tiny bits of DNA gathered at a crime scene with samples collected from felons and suspects. At a moment when the CSI phenomenon is suddenly bringing unprecedented numbers of crime-fighting students to John Jay and The Graduate Center, he has suddenly become quite indispensable.

Kobilinsky, who received his Ph.D. in biology from The Graduate Center in 1977, came to John Jay full-time from the Sloan-Kettering Institute for Cancer Research in 1980, only a few years before forensic DNA testing was about to become a revolutionary discovery for human identification. He has since become a sort of crime-science Renaissance man around campus.

On a given day, he might be advising a graduate student, peering at readouts from one of the genetic sequencing machines housed in his compact laboratory, writing a textbook on forensic science for high school students, or appearing on television to comment on a high-profile case of the moment--Kobilinsky was prominently featured for months in a seat beside Dan Rather, analyzing the scientific portions of the O.J. Simpson murder trial. A recent week even found him jetting off to the Dominican Republic to investigate that nation's penal system and write an eye-opening report at the request of newly elected President Leonel Fernandez, a City College graduate.

Back home, Kobilinsky's own criminal justice research takes several tacks. One is his recent push to add greater rigor and science to the way crime investigators collect, document, examine, and match bite marks to exemplars (orthodontic casts taken from suspects).

"Why bite marks? Virtually all Federal grant money is going into DNA testing now, but that has been to the neglect of other kinds of evidence--latent fingerprint, hair, bite-mark, document examination, fiber evidence, and other kinds," he explains.

Indeed, in a number of recent cases, fingerprint or bite-mark findings of guilt have clashed with subsequent DNA test results exonerating the same suspects.

"I've been very critical of forensic procedures that are more art than science," he says. "So much of this bite-mark examination is subjective, just as it is with fingerprints. There is no clear protocol or system for the examiner who must make the call as to whether or not this bite mark matches the bite pattern from a suspect; it mostly comes down to feeling, opinion, and experience."

With graduate student Kathleen Pfeiffer and forensic dentist Dr. Ira Titunik, Kobilinsky has been using an image processing computer software known as Lucis to digitize, and then transform into grey tones, photographs of bite marks on victims. The resulting images can then be matched to a growing catalogue of Lucis-transformed images of sample bite marks from suspects.

"This is about putting more objectivity into that examination," he says. "What we're trying to do is bring more science to that process."

Kobilinsky also remains active on the DNA front: Another of his projects is a concerted effort to improve the scientific protocols used in lab work. New advances in DNA testing, particularly of so-called low-copy number DNA samples, have increased the power of the tests but also the potential for errors in identification. Because the low-copy method uses such small numbers of DNA molecules--genetic profiles have been generated from as little as a single human cell (and there are trillions of cells in the body)--the testing must be done especially carefully.

"These low-copy techniques are very useful in cases of burglary and robbery, where you might only find a single smudged latent fingerprint or a bit of hair," Kobilinsky says. "You might turn up some good investigative leads. But the flip side of this is that contamination can ruin everything. For example, if even a few cells slough off the investigator's hair or skin and gets into the sample, it's contaminated. Or if some DNA-containing evidence was already present at the crime scene prior to the criminal event, you could generate an irrelevant genetic profile. Even if you get 'successful' typing results, you may not always be able to interpret them properly."

With his John Jay lab team, Kobilinsky is devising a new system for handling small amounts of DNA and amplifying that DNA from a few copies into billions of copies using the process knows as polymerase chain reaction. At any stage of processing (extraction of DNA, quantification, addition to PCR reagents, or thermal cycling) it's possible to accidentally introduce a contaminant which could render the results difficult or impossible to interpret. To test the refinements, he obtained nearly one hundred old-case weapons swabbings from a collaborating police department (the John Jay labs are not accredited to do any actual crime scene lab work for law enforcement purposes), then re-ran the DNA testing using a number of different protocols.

"I feel we've hit upon a better way" of organizing and performing the low-copy tests, he says. "The technique we use minimizes exposure of the template DNA to any potential contaminant."