Supria Rosner

You grew up here in LA, but left for a while. How did you end up back here?
I did my undergrad at Berkeley to study science. Everybody was really interested in me being a doctor, but I always thought about forensics back when I was little. Actually, I saw this movie when I was eight or nine called Feds with Rebecca De Mornay, about women FBI agents. I've always liked science and math, so I thought, "Okay, this would be one way to merge them."

Back then, there was no [TV series] CSI or anything. When I graduated high school in '96 and went to college, there were only a few schools with forensics programs at the time. One of the forensic schools started in Berkeley, but it shut down. A lot of the former students went on to create new schools—one of them was Dr. Peter De Forest, who started a program at John Jay College of Criminal Justice in New York.

When I was an undergrad, I was like, "Oh, I'll never be able to go to New York." I grew up in California, and New York was always so far for me. But it just so happened that my sister moved out there to become a legal-aid attorney, and she said, "Oh, my roommate left; do you want to move here?" and I said, "Okay, I'm there."

Then my acceptance into the program was delayed. My professor had gone to Berkeley, so I kind of gave him a sob story. I said, “Please, I've been wanting to do this since I was seven; I don't want to wait a year!" So then, he said that they’d accept me on a preliminary basis and see how I did.

That's amazing!
Yeah, but the program was very small. I graduated college in 2000. There wasn’t a lot of exposure about DNA testing other than the O. J. Simpson case, which kind of introduced the world to DNA testing.

When I graduated, I got a job at the Office of Chief Medical Examiner, the OCME. They had a forensic lab that just did DNA, and that's where I first started.

Why DNA?
I think because my background was biology. There are a lot of different forensic disciplines; I did an internship at the NYPD where they did a lot of drug chemistry and things like that, but I always had an interest in DNA.

I was a DNA analyst at the OCME, so I was analyzing homicides, property crimes, sexual-assault kits for biological fluids. What was big about New York is that the city did most of the World Trade Center victim identifications. I started at the OCME in '04 but we were still doing identifications from 2001.

They spent years and years going over the 20,000 fragments that were found at the site. Some pieces are smaller than the tip of a thumb. It's a huge project and it's still ongoing. When I was there, they had created a sifting facility because they realized that all the dirt that they removed right away still had some fragments in it, and so they sifted through all of that dirt.

Cases like that are difficult because when there’s a lot of heat, there’s not a lot of DNA left. So you often have to do other types of testing to see if you can identify it.

There are four steps to the DNA process. The first step is extracting the DNA, getting it out of those cells and then cleaning it up. The second step is telling you how much DNA you have; you need that to optimize the further steps to generate a DNA profile.

Ninety-nine percent of our DNA is all the same; it's just a small percentage that's different. Our DNA is filled with these repetitive units—it's the number of these repeats that's different between person to person, and that's what we look for, forensically, with DNA testing. If you see a DNA profile, it looks like a string of numbers, and we look at locations on the chromosomes. You can kind of think of them like addresses on the chromosome. You have two numbers, one from your mom, one from your dad—that number represents the number of repeats, or that repetitive unit in your DNA, so that's why you get this string of numbers.

If I say that at this one location I'm an 11-12, that means I have 11 repeats on one chromosome from one parent and 12 from the other. Then, I obtain a DNA profile from blood or biological fluid from a crime scene, and compare it to a known person to either include or exclude.

So, there’s a giant database of DNA profiles that you can search?
There's an FBI database that houses a group of known individuals, mostly convicted offenders. The database has different sections: one is unknowns from DNA profiles, another one is convicted offenders or known individuals, and the third is a missing person's database.

There are a lot of rules for what can be compared. But let's say your car gets burglarized and the suspect just happens to leave a cup behind. We would be able to swab it, type it, generate a DNA profile, upload it to the database, and see if there are any potential hits.

What about when you’re looking for someone who doesn't have their DNA recorded anywhere?
Yeah, so sometimes those are cases that are unsolved. An investigator could say, "I have an idea for who it is, so let me get you a reference sample from that person so you can do a comparison." Not everybody's in the database, so then we can do a case-to-case comparison, or sometimes we obtain a profile. The system is always being updated with new profiles.

California became an “all-arrestee” state, which means if you're arrested for a felony or a misdemeanor, you have to provide a DNA sample that goes in the database. Not all states are like that, though more of them are moving towards that. And then, if you're not in the database, there's something called a familial database search.

There was a really big case that I worked on called the “Grim Sleeper, which was a serial pattern of homicides dating back to the '80s. Over 20 years, there was always this DNA profile that was found on every victim’s body. He was called the Grim Sleeper, because he “slept” in between the murders; there'd be six or seven years of no killings in between.

We always had the DNA profile but no match. Eight or nine years ago, the police still didn't know who the killer was, so they created a task force to see if any work could be done. So finally, the Department of Justice and the police department worked out a way to search the database for not just a specific person, but for a relative of that person. This was very controversial at the time—because if you see DNA that's somewhat related, it might not be from an actual relative, it might just be a fortuitous match. You have to do your investigations to figure out. California was one of the first states to do it.

In 2007, they decided to run a database search to see if they could find any relatives of the killer.  Sure enough, they found a relative in the database; it turned out to be the Grim Sleeper's son, who was arrested for a felony weapons possession.

Now they investigate: Does the son have any relatives that could be the suspect? The investigation pinpointed this one person, so they started following him. After a few days of not leaving his house, he went to a pizza party in Orange County, so our surveillance unit followed him and pretended to be a busboy. They collected this guy’s pizza leftovers and brought them into the office, and that's what I got.