JURI Committee Meeting

STANDING COMMITTEE ON JUSTICE AND HUMAN RIGHTS

COMITÉ PERMANENT DE LA JUSTICE ET DES DROITS DE LA PERSONNE

EVIDENCE

[Recorded by Electronic Apparatus]

Tuesday, March 10, 1998

• 1537

[English]

The Chair (Ms. Shaughnessy Cohen (Windsor—St. Clair, Lib.)): This is Bill C-3, an act respecting DNA identification and to make consequential amendments to the Criminal Code and other acts. We're also reviewing Bill C-104, which is on the warrant sections of the Criminal Code concerned with DNA testing.

Today we have with us Dr. Ron Fourney, who is the research scientist in charge of DNA methods and database at the RCMP Central Forensic Laboratory. Welcome, Ron.

Dr. Ron Fourney (Research Scientist in Charge of DNA Methods and Database, Royal Canadian Mounted Police Central Forensic Laboratory): Thank you.

The Chair: We know you have a presentation and a slide show.

Dr. Ron Fourney: Yes. First I would like to thank the committee for the opportunity to talk today. I've been monitoring the various comments and questions which have arisen during this committee. What I've done today is going to try to address some of the technology, I hope in a simplified form, so folks will be able to understand it, and perhaps put to rest some of the questions that seem to be recurring on a routine basis.

I do promise I'll make this a bit easier to understand than my presentation yesterday at the University of Toronto. If there are any technical questions, I hope people will ask me about those. I'll try to explain.

The Chair: You will have to go slowly today, because it's mostly lawyers.

Dr. Ron Fourney: My section is DNA Methods and Database. In the RCMP we're a relatively small group. There are five of us. We are essentially in charge of a very important aspect of the DNA technology. It is our responsibility at the RCMP to develop, test, and validate all the DNA methodologies before they are actually given to the operational groups for analysis. Consequently we deal with a lot of the troublesome parts of the technical side. We get a procedure to work properly so it's valid and reliable.

Today I'm going to start off with the history of the technology and what we have done. First of all, I would like to point out that as a forensic scientist, I'm also a molecular biologist.

This particular picture shows where DNA started. On November 21, 1983, about 15 feet up on the side of this black pad in Enderby, Leicester, England, a 15-year-old, Lynda Mann, didn't come home that night. In fact, she was found murdered. She was strangled and sexually assaulted. When they did a vaginal swab they found, from a protein technology that was currently used called serology, it matched 10% of the male population in England. There was a massive manhunt, but they still have not found the killer of Lynda Mann.

• 1540

Three years later, less than a five- or ten-minute ride from the spot at another small dark pad called Ten Pound Lane, Dawn Ashforth was sexually assaulted and brutally strangled. When they did a serology analysis of her vaginal swab, it actually matched the first victim as well. Consequently, the police knew they were looking for someone who potentially perpetrated both crimes, and 10% of the entire population of England would have been a match.

About this time, a young geneticist working in Leicester, England, on evolution discovered a piece of DNA that seemed to be highly variable between different people. That person, Alec Jeffreys, later went on to become Sir Alec Jeffreys. He was knighted for his work and he is essentially the father of DNA typing.

The police brought the exhibits to Alec Jeffreys. They processed them and he had two things to tell the police. They had originally picked up a young kitchen porter, Richard Buckland, who had confessed to the second murder. He knew the victim and confessed to the crime. When Alec Jeffreys matched his pattern with that of the other patterns, essentially they did not match. The police were told they had a criminal who had committed both sexual assaults and murders but had picked up the wrong person.

Essentially, in 1986 legal history was made. The very first time DNA typing was ever used from a criminal point of view, it was used to exclude someone. So we're talking about exoneration of the innocent.

There was a massive manhunt in the vicinity. They bled 4,500 men within a certain age group. On the first pass through, they actually missed the accused, Colin Pitchfork. Someone else had given a blood sample in his name. It was only during a casual conversation in a local pub that someone overheard that someone had given a sample for Colin Pitchfork. They went out, bled Colin Pitchfork, and essentially had a perfect match. This is all described in Joseph Wambaugh's book, The Blooding. To me, as a forensic scientist, that is the first real situation where DNA typing was identified as a powerful technology.

I'd like to point out that from a forensic DNA typing point of view, it's highly discriminating. That is why forensic scientists like the technology. No two people have the same DNA pattern except identical twins, who are essentially clones. Genetic continuity simply means your skin cells, hair, blood and semen will contain the same DNA. It means the DNA you were born with is the DNA you will die with. It's highly sensitive. You need very small amounts.

You'll see today when I go through the history of the DNA typing that we started with blood about the size of a quarter and we're now down to what will fit on the end of a pin. We don't need very much material. Probably one of the key features to forensic DNA typing is the stability. This is a unique molecule. Deoxyribonucleic acid is not like a protein that tends to break down when it's challenged by environmental insults or bacteria. This stuff is rugged. This molecule withstands many types of environmental insults and can be used for typing very old material. So you can match a case that's 17 years old, for instance, with a case that occurred yesterday.

The source of forensic DNA typing is blood, semen, tissue, hair, saliva, tooth pulp or bone. Essentially all these materials would give you the same genetic DNA pattern. This is basically what we are interested in as forensic scientists—material found at the scene of the crime. It's once again associative evidence.

The double helix is a molecule with two types of helical ladders. There's a very simple genetic code here. Unlike our alphabet, which has 26 letters, the genetic code has only four. It's this repetition of the four letters in a sequence that makes a difference between me, my brother, and you in the audience. If the sequence is slightly different, then the person will be different. If the sequence in this ladder is the same, then you either have a sample from the same person or you have a genetic identical twin. This is what all the fuss is about. This is the deoxyribonucleic acid. It's been purified from all the cellular membranes, proteins, and walls of the cell, etc., and been broken down. It's a close-up shot showing that when we precipitate it out using a chemical it looks like a white, flocculent long piece of thread.

• 1545

In reality, we're in the business of human identification. I'd like to think that these two young FBI agents at Quantico have a complete complement of DNA in their systems. Let's have a look at them in a close manner. If in fact they're brothers, they're going to share bands between themselves. For instance, here's mom and here's dad. One band from brother one shares with mom, a second band shares with dad. This tells me a lot of information. It tells me that they're related. It tells me that they're different and that I can identify these as individuals if I do the genetic testing with enough series of tests.

Let's have a look at this. This is what we call an RFLP DNA typing membrane, which I did way back in 1989. I think it was in 1989, a year after I joined the force. And here are some ladders. These are just molecular weight markers; these are simply like a ruler. And here's mom's band. There's mom's band here matching child one and dad's band matching child two.

I'm particularly proud of this membrane for a lot of reasons. First of all, this whole sample here was done with a buccal swab. That's simply a toothpick on the inside of your mouth. We extracted the DNA and that's what you have. This was done with a single strand of hair. We extracted the DNA from the root and we got the same pattern that matches. These happen to be my children. I was quite pleased that the paternity of course worked out fine. You'll see that Christopher here matches me and matches my wife Anne here. So there's a biological identity here. It's a Mendelian pattern. It's very sensitive. It's genetic continuity. This is the gold mine of forensic DNA typing. It allows us to identify individuals.

Just when we were getting complacent with the technology, in 1992 a brand-new revolution occurred. I would like to think that there were two types of revolution in the forensic field. The first one occurred when we discovered the DNA and applied it to forensic. This was akin to discovering fingerprints over a hundred years ago. It was the most significant human identification tool to come around in one century. And then a short few years later came something called the polymerase chain reaction, PCR. I'm going to show you how that works. It's a complicated procedure, but I think we have it worked out so that you'll understand. This is a technique employing a specific way of looking at DNA that even makes it a much more powerful technology for forensic and also for basically clinical analysis and diagnostics, etc.

Why are we interested in PCR? It's very sensitive. We need smaller amounts. It works great on degraded material. So as DNA gets older and becomes more degraded it gets smaller; there are pieces of DNA. Sometimes these are hard to work with with that first procedure I showed you. So this PCR technology can work with smaller amounts. It's amenable to automation and rapid processing—and I'll show you how that works in a minute—and through automation and rapid processing we can add all kinds of additional quality assurance standards.

So this is a very sound technology, and Kary Mullis in 1993 won the Nobel Prize for it. It's probably the most significant technology that has come into science, I would say, in the last three or four centuries. It's a very specific, high-technology tool.

Where is PCR used? It's used for identification of individuals in airplane crashes. When the TWA 800 plane went down off Long Island a few years ago, the technology that was used to identify the bits and pieces of the various individuals who were killed in that massive crash was the PCR technology. Jack Ballantyne in the Suffolk County lab employed a very similar technology to what we currently use at the RCMP labs.

There's a technology that was used in Desert Shield and Desert Storm to identify those killed in action in the front lines. It was probably one of the first times in armed conflict in the history of man when they could identify the casualties almost completely in every situation. There was one particular armed personnel carrier that was hit by a Hellfire missile, shot from an Apache helicopter. There was complete incineration. There were five people in that armed personnel carrier, and essentially they were able to identify every single one using the PCR technology.

• 1550

It is also a technology that allows you to look at ancient remains. A few years back, a couple of Swiss tourists were climbing up a mountain. They looked over and saw this body coming out of the snow. Now, I don't think this was a homicide, but they do estimate that this person is about 4,500 years old. This is Ice Man, who became a significant anthropology find at the time.

We're now looking at the way man has evolved over the last few thousand years and how their gene structure has evolved in that time. We have forensic scientists working in conjunction with academic people who are looking at the history and evolution of man throughout the ages.

It was also used to identify the remains of the royal family of Czar Nicholas II, who died over 70 years ago. One night his entire family was shot and put into a grave site that moved several times. The remains were burnt, I believe, and acid poured on it. The British Home Office forensic science service used PCR technology to identify the remains. I think you'll find that he is being buried now in his proper site in Russia. It was quite a publicized story. It's PCR technology that allowed us to do this.

From a forensic point of view it's a very powerful technology. We used to work with about 50 microlitres of blood. A microlitre is about the size of a quarter, say. PCR technology allows us to do things like postage stamps or envelope flaps. We can swab a telephone receiver. We can look at the blood splatter here, or we can look at the dots up here. That is the generation change in the technology, the sensitivity increase, that made us very excited about using PCR. So to have two revolutions in forensics within a decade was incredible.

This is a very complicated picture. I'm going to make it simple for you. This is the polymerase chain reaction. All your chromosomes containing your DNA are wound tightly into 23 pairs, or 46 chromosomes, into your cell nucleus. What we do essentially is unravel these pieces of DNA into the tubes of ribbon-like alphahelic components.

Here are the two strands. With the polymerase chain reaction, all we've done is use nature's natural chemicals, enzymes, to go in there, break the DNA apart, and if we add linkers in here, called primer sites, and we turn an enzyme, called the polymerase, onto it, and add all the building blocks, you get a new piece of DNA that is essentially faithfully replicated from the original target sequence.

If you do this one time, you get two copies. If you do it twice, you get additional. If you do it thirty times, you get over a billion copies. Essentially, this is the whole secret behind PCR technology.

What we like about it is that if we add different-coloured components, called fluorescent labels, into one tube, we can do multiple analysis simultaneously. The old RFLP technology used to take us anywhere up to eight weeks to do a test that was composed of the various components. In today's technology, this can be done in days.

The secret behind it is the heating and cooling of DNA to allow the enzymes to go in there. So if we add in all the components and put it into a very sophisticated heating and cooling machine, a thermocycler—it's like a molecular Xerox machine—it would go on and make additional copies, faithfully replicating and adding a colour component tag to each piece of your DNA.

There's your molecular Xerox machine, the 9600 thermocycler we use in the lab.

Here's the procedure we're currently using. We search the exhibits, we extract the DNA, and we clean it away. It looks like that filamentous piece of DNA I showed you at the beginning. All that's in the tube is DNA. We amplify it with our molecular Xerox machine, the thermocycler. We put it onto a very special machine.

I'm going to talk a little bit about this, because I think some questions arose regarding how we detect it. Because the pieces of DNA are labelled with fluorescents, we're able to use a very sophisticated laser in a camera to pick up different colours of DNA. We can order these. When we know the actual order of these, we can identify the different components, linking it back from one person to another.

How much DNA do we have? This gets into the question of, how are we unique? Well, let's assume we're talking Lego blocks here. This is a presentation I prepared for the grade six class at my son's school.

• 1555

An average piece, your basic component of DNA—we'll call it your Lego block here—if we take that as a three-centimetre piece of Lego, we have enough DNA in one cell in Lego blocks to cover 90,000 kilometres. That's the entire coastline of Canada. If we blew up our single base pair component of DNA to take into account the fact that we have three billion, it would cover the entire coastline in Lego blocks.

What we're actually using, though, is only 0.1% of all that 90,000 kilometres. In other words, 90 kilometres of that is different from person to person. So in actual fact we may think we're different, but we're more alike than we are different. And in that 0.1% there are some very nice areas of polymorphic regions, which we forensic scientists will use, because those areas are particularly polymorphic, that is, different, and allow us to do identification.

All this technology is coming out of a major project—I think this was in the 1990 issue of Science—called the human genome mapping project, in which scientists all over the world are combining their efforts in order to actually sequence the entire human map so that we know all the different components. And the reason this is important is that they're looking for linkage sites that will perhaps tell them if a person has a particular disease or a predisposition for a disease and so allow for treatment.

Yesterday I was at the University of Toronto. I talked at the Hospital for Sick Children. One of the groups there is actually just sequencing and looking at one chromosome, chromosome 7. I was interested in sharing our results with them in the sense that some of our techniques have gone full circle.

When I first started in this program, I worked in cancer diagnostics and genetics, in 1988. Then I developed an interest in the forensic field, and I was asked by the RCMP to come into the program and meet with their team in order to develop a DNA program. So at that time, we took from academics, research and diagnostics and used that in forensics, and now it's going back full circle, because with respect to some of the procedures we worked out in terms of validity, reliability and sensitivity, the folks in the clinical diagnostics also want these procedures.

Here's a chromosome map. You have basically 23 chromosomes. This example happens to be a male. It's a Y chromosome here. It's a little bit smaller. It's actually a fairly interesting chromosome because it's a fairly boring chromosome. Most of this Y chromosome has a repetition in it that's not very different. As a result, the X is more interesting for us to look at.

Some hon. members: Oh, oh.

Dr. Ron Fourney: Nonetheless, from a sex-typing point of view, if we can tell the difference between a male and a female it's very important at the scene of a crime.

Each one of the yellow and red spots are interesting areas of variation that we, as forensic scientists, want to look at. And the interesting thing here, too, is that we're looking primarily at different chromosomes. The red ones are the new technology called PCR, and the yellow are the old technology we used to call RFLP, restriction fragment length polymorphism.

Why are we looking at different chromosomes? It's sort of like the odds game in trying to determine discrimination. Let's pretend this isn't coding for a particular piece of DNA but in fact is coding for a characteristic of somebody. Let's say this chromosome 1 codes for bilingualism and chromosome 2 codes for brown eyes.

If I know these two areas, just as a characteristic, I could say that I can start discriminating people here because they're bilingual and they have brown eyes. And then we could go for a really rare trait—let's call it being a hockey fan of the Toronto Maple Leafs. People say that's rare for me, but I happen to like the Leafs, and let's say that one in 2,000 people are like that. Let's say that's 50% of the people, one half, and this is one half over here. That's one half times one half times one in two thousand.

And those are your odds. Because they're all on different chromosomes, we can multiply those together and determine the identity of a person based on discrimination, because each one of these is a separate test.

Let's look at this in detail. There's one thing I want to point out, because I know you had one witness here who was quite concerned that we could identify other parts or other features with regard to the typing that we use in forensics. That's not true. At this point, we're only working with what we call “anonymous pieces of DNA”. We don't know what they code for. We're not looking at blue eyes. We're not looking at a trait for schizophrenia. We're looking at pieces of DNA that essentially have no genetic coding region.

• 1600

In fact, we push this even further. I'm a member of several world committees dealing with the identification of forensics, and we work cooperatively. In 1993 the International Society of Forensic Human Genetics Conference actually talked about this and we came up with the conclusion that we have to carefully choose a set of DNA markers that do not code for any known genetic disease or information of any predictive medical value.

This has been done in cooperation with leading experts in the field of medical genetics and human population genetics. In fact we routinely publish our work, and we discuss it openly in conferences. We talk to people—like yesterday at the University of Toronto—and we're constantly aware of new developments, because the last thing we want to do is use an anonymous piece of DNA that all of a sudden codes for a genetic disease. If that happens, that marker should be removed from the database.

To date, these have been highly and carefully selective. We have had the authority of numerous population geneticists working with this and it's a cooperative effort worldwide to actually zero in on the short tandem repeat, STR, markers that we're currently looking at.

I will try to explain this to you, because it's an interesting subject. I'm going to explain how PCR works and what we're looking at. Let's say volume 21 was chromosome 21, a particular area called page 27 and a smaller region called line 18, and you happened to get this sentence from your father: DNA is a long, long, long molecule that is tightly wound. If PCR is a primer site, this is that region I told you about where we're binding other pieces of DNA, and we're going to make a stretch here, as in the red, and it's identified by the word “is” and you have this second sentence from mom, you only have two stretches of the word “long” in here and you have four in here.

What we're going to do is look at the differences between the sentences by looking at the length differences in DNA. How many stretches? This, believe it or not, is called a short tandem repeat. We're using words and sentences here, but in fact the genetic alphabet is just composed of four letters: A, G, C and T. So that's what would be here, and that becomes an STR.

Let's say we look at another chromosome, chromosome 6, a particular area, page 2 at line 128, and you happen to get this sentence: DNA typing can identify, identify, identify victims of mass disaster and can be and so on. Mother, DNA typing can identify, identify. Well, this is a totally different page in the book, volume 6, volume 21, chromosome 21, chromosome 6, and those are the short tandem repeats there and this is the region we're going to hit with our polymerase chain reaction to make more copies.

So now we have two different sentences. How do we actually look at these? Here are all your sentences here, but we have a way of— Once we do the polymerase chain reaction it makes only the stretch in between here.

So between the two PCR sites, the word “is” in this particular case, right there and right here, becomes your short tandem repeat. We use a gel electrophoresis procedure. Think of Jell-O poured between two glass plates. You put the DNA on top of it and you run an electric current through it. Well, DNA happens to have a negative charge and it's going to go from negative to positive and all the smaller fragments are going to move much faster through this Jell-O. Essentially, what you're going to have is a way of separating all these different-sized fragments, short tandem repeats, from long to small.

That's the whole principle of what we're going to be looking at for STR analysis for forensics.

Let's say we throw a molecular ruler in there. I know the sizes up here. If I know the sizes of this, I can relate it to all the different components. If I see red here, I know the size of this one is larger than that one and I can basically start figuring out the fragment length. We're essentially measuring the number of repeats in this region from the large to the small.

This is what it looks like. This is your gel electrophoresis is one test, green is a second test. Now, remember I told you about our sex typing. This happens to be a person with a Y chromosome, so he's got this band and he's got the X chromosome. This happens to be a male. Here's a female; it has two X chromosomes, missing a Y.

What I want you to remember is that this— One lane down here is one person; this is another person down here. There's a third person down here, and we're running this simultaneously altogether. This is very important. And you'll see in a few minutes why you can't take just one lane out of here without destroying the entire picture.

• 1605

What is the likelihood that someone other than the suspect was the source of the evidentiary material? This is essentially the question. If we don't get a match, it's an exclusion—the game is over; it's an exclusion. If we get a match, they want to know how many people in the world have the potential of matching that component. If we just do one test, chances are a lot of people are going to match. If we do two tests, fewer people are going to match.

What it works out to is currently in the RCMP we run three multiplex systems. That's three different large series of tests: this one has three different STRs on it, three different tests; this one has four different tests; this one has three more tests and the sex-typing test.

The average discrimination—we just ran this test—is one in 7,812. So we can discriminate 7,812 people. Basically that's what this means, the potential discrimination. This one is one in 9,345. This one is one in 1,275.

The combined discrimination using all those tests is one in 93 billion. So this is essentially the power of the technology.

One of the problems we have is explaining in a court of law how we know we can exclude people or how we know we can match people to a certainty.

If we just look at our first test, called multiplex 1 and multiplex 2, the most common in the Caucasian population is around one in 1.2 million. That's not saying it's the most rare; the most rare is 2.2 followed by 31 zeros. How do you explain in a court of law how rare that is? We had all kinds of recommendations. One of the ones I like is that in 1995 the Ice Research Centre in Canada estimated that 10²⁴, that is, septillion, snowflakes fell in Canada annually. Our tests are getting to the point where you may become more rare than the entire number of snowflakes that fall in the course of one year.

Now I'm going to walk you through the history of the technology, and this is important, because I think one of the questions was why it is important to retain the sample in our data base.

How quickly does this technology change? As a molecular biologist, I can tell you that it is probably the most active, challenging, and quickly changing field of any of the sciences. When I first started in 1988, this is the type of DNA typing I used when I worked on breast cancer and colon cancer. When I joined the forensic group we developed a procedure that made it a little bit easier. You could imagine trying to go into court with this and saying there are two people in there. We've simplified it to just a series of bands.

In April 1989 we became the first police forensic lab in North America to use their own DNA typing in court. I believe the FBI used their DNA typing about three weeks later. Up until that time, most forensic services were being rendered by commercial companies, private companies. So in 1989 we went into court, and that's what our DNA block looked like. There was a sexual assault on an elderly lady. I'll just run through it quickly to show you. The accused is here, two bands. It matched the semen found at the scene of the crime. And these two bands were the actual victim sample.

When we did the calculations across seven different tests, I believe the chances of matching someone in the general population was in the upper billions. When we arrested our prosecution side, there was a short recess. When we came back, the defence made a motion and basically the accused pleaded guilty on all counts. That was our very first case, in 1989.

Let's look at how fast the technology has changed. That was 1989. A year or so later this is what the gels looked like and the typing. You see how the markers are quite different.

It is an interesting case. I'll quickly tell you what happened here.

These two fragments of DNA match these samples over here. This is blood found in the accused's car, because he moved a body. This young woman was killed, blood was all over the place in his car trunk, and the blood was found in his car. The DNA of the body was found in a dumpster. What had happened was a body had been placed in a dumpster, gasoline was poured into it, and it was set afire. The incineration temperatures inside that dumpster were so high that it rendered the body of a young female into about three or four pounds of material at the bottom.

• 1610

In the jaw of the body of the remnants in that dumpster were one or two teeth. We went in, took the bone marrow out of the teeth, and we matched it back to the blood scenes found in the accused's car. That's quite an interesting case. It shows you how incredibly stable the DNA is, to withstand the temperatures it did.

Then in 1991, this is the technology; the first time we used a PCR—not very exciting. It was giving us results. We went to a colour process—a bit nicer. This is about 1993 or so. We're starting to get better resolution.

This is what we're using for the database today, if we go ahead. This is the state of the art. It was commercially released just in December. We've been working on it as a field-testing validation site for this technology for two or maybe three years.

One system now is run in one tube. It does nine tests and the sex typing. That happens to be a male there. You see the X and the Y chromosomes, two bands. These are all females. What we used to take three gels to do, and three consecutive multiplexes, we can now do with one.

In June we're going to begin training our operational people with this technology and we're going to convert completely.

What you have seen is an evolution from 1988 to now. That's now much the technology has changed. And it's going to change further. As part of the requirements of my program, I'm always looking two or three years ahead of time.

This is the entire mitochondrial DNA—that's another type of DNA found in the body—put on a microchip. Every time you see a little band, it means a sequence has hybridized. Essentially you can get a result in about five seconds with 16,000 or so base pairs of DNA. It's all on a chip about a centimetre by a centimetre. It's not quite ready for forensics, but I think you will see a diagnostic approach used with microchips in about three years.

Forensic is always the orphan child of science, in the sense that if a laboratory wants to make money commercially, it will develop a diagnostic test first. Then the technology will be modified for the areas we're interested in. That's how forensic will get this chip technology.

This is where we were in 1989, running a gel that was loaded manually on an Agarose type of material. That's Jeff Madler, from our Halifax lab.

Today a 377 sequencer costs about $150,000. It automatically runs the DNA through. A laser goes back and forth, scans it, and you can see the image coming up on it. We have twelve of these sequencers in the RCMP laboratories. There are four in Ottawa, four in Regina, and four in Vancouver.

I spoke at a meeting in October where they wanted to know how fast DNA typing had changed. They asked me, as someone who has been in it for a long time, to bring this up, because I can probably remember the older techniques.

I went back through the history of the International Society of Forensic Human Genetics. I looked at all the proceedings published from their meetings in 1987, 1989, and 1991, and I calculated the percentage of talks given on the various PCR. In 1987, 94% of the talks of this large European, international meeting were based on RFLP technology. Today 97% of the talks are on PCR. Hardly anyone is talking about the older RFLP technology. That's from 1987 to 1997.

Really to rub it in how fast this is changing— when I had a Sunday off a couple of years ago I went to the British History Museum. There was the first thermocycler I had used as a student, on display in the British History Museum, showing the history of molecular biology. Something I remember using in the 1980s is now considered good enough to be put on display in the British History Museum.

Let's look at how the technology works in casework. This is a situation where I'm going to describe what a typical sexual assault would give us with the technology.

Here's an indication, number 7 here— these are three different fractions of DNA. When we get a vaginal swab, by using what we call “differential extraction” we can actually isolate the male component from the female epithelial cells. So we can take the DNA that is found in the sperm and separate it from the DNA that's found in the epithelial cells of the victim.

That's what we've done here. The victim, number 10 here, has this profile. Number 11 is the suspect. What we did with the first processing of the sample will render the female epithelial cell—see how that fragment matches her. These two match her, the second two match her, and these two also match her again.

• 1615

We go through several different processes, and we end up with the male fraction that does not match the victim, but certainly does match suspect number one. Suspect number two here was excluded. That's one way it works from the gel technology.

As forensic scientists, we like to read our pictures and graphs like this. We can print them out and get this kind of fragment analysis.

The same thing's happening here. This is a male. It has the X and Y chromosomes of suspect one. Here's suspect two's X and Y. And here's a crime scene sample. You can see that these don't match. There's an exclusion here and another exclusion here. They match perfectly with suspect one's.

Here's how the technology actually works at the crime scene: We'll collect samples at the crime scene—in this particular case, there are four of them. They come into the forensic lab and the case file is opened. The evidence is carefully examined, and they fully describe and document all pieces of the evidence. They do some initial tests to see if the samples are blood, urine, or whatever. The sample is then sent over for DNA analysis with our 377 sequencers, and you get this raw data file that we call an STR gel file. It has to be processed with the computer to give us the information that we'll actually use.

See the graph here? It simplified all this into a graph, and what we can in fact do is actually simplify this even more. This whole pattern of one sample going down here can be simplified into a series of two-digit numbers that relate to each one of these fragments based on the number of repeats and the size of the fragment. For instance, this lane here could be 12-14, 11-18, 20-28—it's a male X-Y—11-16, 27-34. That is the final DNA profile that we are interested in comparing.

Your raw data goes through computer processing and it's looked at in different ways. We can look at it in this graphical format, which is a simplified format. In reality, however, it comes down to something that looks like this. This is the final DNA format. It becomes important, because one of the questions that was asked was what we are actually storing and what we can cut from the gel, or how we excise information from the gel.

Here's suspect one. It's X-Y for suspect one. It's a particular test; it's number 10 and it's number 12. The same suspect one has a 29 and a 30.2, etc. At the crime scene here, you can see the D21S11. That's a test here on chromosome 21.

I'm going to teach you how to read chromosomes. DNA— D stands for chromosome 21. S is a single-copy gene, and it was the eleventh time somebody reported a particular site on that chromosome. So you're all unofficial molecular biologists now. So 29 matches with 29, 30.2 matches with 30.2, etc. This is the final digital code.

The national DNA database will be an investigative tool. It only points the police in the right direction. It can link crime scenes together or it can direct the investigation towards a criminal offender. That's the investigative tool many of the police departments really want, both across Canada and worldwide. They want to link the crime scenes to a particular person.

I want to preface my talk, actually, to indicate that we don't know exactly how the DNA database will work. We're still in the discussion phase with our partner forensic labs in Canada. I'm going to suggest to you one possibility of how it could work.

Some of our guidelines will come from you folks based on the regulations and the legislation. For instance, in this particular case in which we're processing DNA evidence, the information that will be sent here will include a bar code, an alpha-numeric number, the submitting agency, the casework file reference—just a file number referring to that particular sample—the date of submission, and the simplified STR pattern. This is what it will look like. If the Centre of Forensic Sciences was sending us a sample for analysis in the database, it would probably look like this: ACX56783. That's a unique identifier. It came from the centre.

• 1620

That's their gel file. We don't know what's in their operational case file, but that's their operational case file. That's important for them to know so they can relate this back to it. The data was submitted, and they'll be sending us a series of bar code numbers. That's it; there's no other information in there.

This is a busy slide so I'm going to probably skip over it a little bit. But here's what they're sending. Forensic Sciences is going to say we should compare this sample with a criminal offender. We will go through it. Let's say we find a match with a criminal offender's sample. Then it's reported to the criminal history file, and it's up to them to notify the forensic lab that they have a match and who it belongs to.

We don't know; we only have a bar code and this series of numbers. We don't know who that match is. The identity is in the criminal history file.

Similarly, with the crime scene index, let's say one sample matches another one from our RCMP lab in Vancouver. All we know is that the two samples match. I don't know any specific information about that other than the fact that they match. I can tell where they came from and when it was submitted, but I don't know if it's a suspect's sample or if it's a blood sample. I don't know any of the relevant information that's important to the operational forensic laboratory to know.

All I'm doing is comparing the two samples and relating back to them. So we'll phone up both of the forensic labs and say that such and such a number matches this number, so we'll ask them to please compare their notes and look at their gel files.

So we cannot provide personal identification on any suspects or other individuals from the crime scene data. We don't have it; it's in the operational files in the forensic lab. We cannot provide any personal identification on the suspected offender because that's in the criminal history file. All we are is a repository of the data so we can compare back and forth.

There was a question a little while ago: why can't you remove one sample from an entire gel image? Well, the first thing I'll tell you is that this is a raw gel file image, so every single sample is linked to its partner's sample. If you take sample number nine out, the whole thing falls apart. I can't relate anything else to sample eleven to sample seven. It's based on the technology we use for collecting the information. This is unique to DNA typing. This is not a standard computer program; this is a very specific program related to a 377 sequencer for collecting fluorescent, automatically tagged DNA fragments.

It's complicated. It will scan 193 times across this gel. That's one scan every 1.5 seconds. Then it will build an entire gel image of 4,500 scans. As I told you earlier, these are bigger fragments than the smaller fragments down here. The DNA actually will move in a linear fashion now.

Here's what we propose. Say someone was in the criminal offender database and we wanted to take out that particular sample. We can't take that sample out without destroying all other 26 samples that are on this gel.

What we can do is this. You remember that the bar code that has come in to identify this sample is here. The bar code going out here is the information that identifies the sample data with which we do all our processing to get our digital code.

Essentially, say we cut the bar code here, and cut that one there. We may have this sample here, but I don't know where it came from or what it is, so I can't process it to get the code. It's essentially rendered inaccessible.

So let's look at this in a little bit more detail. We want to take out sample nine. Well, if I cut the link going in and the one going out, I don't know what it is and I don't know how to process the information, as it's no longer there. The entire gel image is left intact.

Here's why you can't destroy the gel file: it's a raw data file. It's like a filing cabinet that has many files in it: if we take one file out, the entire cabinet is destroyed.

In this machine we have a laser. It's right in this bottom corner, but you can't see it very well. It's a very sophisticated laser that shines on the flourescently tagged material, and the light is emitted.

Behind that laser, there's a camera that captures the fluorescence. It's actually a charge-coupled device. It's the computer that sorts out all the various colours and its actual relationship.

• 1625

So here's the gel if we look at it on the side. We're sort of looking on the edge of the machine. There's the laser and the camera. The pieces of DNA are going to go down here. This is what we call real-time analysis. We'll load the DNA here, and we'll run it in an electric field. It's like a race: this one will get past the laser first, followed by that one. Eventually, all the DNA goes right off the gel into the bottom buffer. We clean it, take the gel out, throw it away, and pour a new gel. So it's like an electrophoretic race.

This is what the company did if they wanted to only analyse one lane at a time. They could put the laser here. In 20 minutes, you're going to see the red one almost go by. In 26 minutes, it's just going by the laser. In 30 minutes, the blue ones are going by. In 45 minutes, the last blue one has gone by. In about 50 minutes, there's no more DNA on this gel. The DNA has moved right off the gel past the laser.

The company was pretty smart: they wanted to not only basically identify one lane, they wanted to identify 26 lanes. Actually, the new technology will allow you to do 96 lanes at a time.

They have a laser that now moves back and forth collecting information. So as the DNA goes past the laser, it's actually moving. Say we pretend we're watching this laser. In 0.3 seconds, it's here capturing that picture. In 0.5 seconds, it's capturing this lane over here. In 0.75 seconds, it's halfway across the gel and catching the image there. When it gets to the other end of the gel at 1.5 seconds, it's ready to go back again. It makes 4,500 scans worth of data. It collects information and goes back and forth. There are 1.5 scans per second.

So what we're doing is essentially like beads on a string with collecting the DNA. This is the first one that's added. You see how the string is connected. This is the second one that's added. This is the tenth one that's added. That's the eleventh because that's the tenth, etc.

Eventually, you put them all together, it gets to the end of the gel, and it starts assembling it as the laser goes back and forth. You do it over and over again, and you end up with a gel image.

So what happens if you try to cut the raw gel file? Basically, you lose all your beads. All the information is gone, along with all the associations with all the other fragments.

In particular, these are your internal-lane standards, these are our rulers that measure all the pieces of DNA. If we try to remove one of these, which the company itself tells us you can't, you destroy the entire file cabinet, not just one particular lane or file of information.

I just want to talk a little bit about the technology. I think someone last week or the week before showed you how we're going to collect various samples. Well, we've refined that even more.

Say we go for bloodstains. We've been working on a technology called FTA analysis. Essentially, this is paper that has been specially treated with non-toxic agents. But when you put blood on it, the cells all break up, and the DNA sticks to the paper. We can literally cut a small piece of the paper out, clean it away from all the blood cell material, and amplify our piece of interest in DNA right off the paper.

This is not a great picture, but some of you may be able to see little, tiny spots. That's a one-millimetre punch. With some of those images I showed you, we can actually do an entire DNA analysis on a one-millimetre punch size of blood. We actually only use 300 of those when we put it on the 377 sequencer. This is a very sensitive technology.

One of the things we do at the RCMP is work in close cooperation with many groups worldwide. The purpose of working groups is to develop conventions so that we're all kind of working in the same direction so we can share data. In particular, we wanted to know what we're actually working on—the alleles themselves—and what we're going to call them. We wanted to standardize the technology. We wanted to develop certain specific PCR systems. We wanted to be able to share databases if the need is there in the future to do so.

We have cooperative validation studies. That means that if a test is run by my counterpart down at the FBI and is shown to work flawlessly and we've run it up in the RCMP, that's considered peer review. People have accepted this, and it's a valid and reliable method. We want to set cooperative quality-assurance standards.

• 1630

I'll give you an example of one of the tests we're doing. In North America many laboratories are embarking on this STR journey. The RCMP lab has been working on this since 1991. We're probably one of the largest labs to work on this technology for the longest time. When we first started, there might have been five labs in the world using STR technology.

Recently, in March of 1996, the FBI, through the Department of Justice in the United States, has funded an international study where we're working cooperatively with 26 laboratories—only a few are mentioned here—to come up with a common set of STR standards we can use so if a sample in Florida is compared with a sample in Canada, they will use the same procedures, the same STR, and they will get the same results. It may surprise you to learn that fingerprinting, the latent fingerprints that are left on objects— in different places in the United States they use different procedures for looking at these fingerprints. What we wanted to do is actually to establish a universal set of standards.

That's what we did. In November of this year, after approximately $1 million was spent on a combined study, with all our labs participating—and once again, the U.S. Department of Justice funded this—we came up with 13 test systems, which will become the international standard for forensic analysis in North America.

I just wanted to show you a bit of the work here. This is a population sample. That's an identification number for a particular sample. Here are all the different tests; and look, there are your data: a 15-16, a 22-24. It happens to be a male, XY, a 12-13. That particular material is what we're actually collecting.

In the offender DNA index, we can essentially cut an entire lane out of here and get rid of it. That's how we could eliminate a sample. If a pardon is given, for instance, and we are told to remove sample such-and-such, we can just go to our database and it's gone. It's not just rendered inaccessible, it has disappeared.

I would like to acknowledge the people in my group. I have five individuals. Over time I have had many students as well. There are Kathy Bowen, Benoît Leclerc, Chantal Frégeau, and Jim Elliott. They worked very hard to develop and validate the procedures we're currently using in the RCMP forensic laboratory systems.

My last slide is just an indication— In July 1993 we had a major paper in BioTechniques. We also made the front cover.

Here's the generation of RCMP technology. It's interesting. I think this went out to 56,000 scientists worldwide. It's the first time I ever sent our family photograph to 56,000 people. Here we have your RFLP typing, a generation of PCR, another procedure we tried, the fluorescent procedure. Even the technology you saw today has improved since 1993.

Thank you for your time. If there are any questions I would be happy to answer.

The Chair: There will be one or two.

Mr. Ramsay.

Mr. Jack Ramsay (Crowfoot, Ref.): I found that very interesting, but probably beyond my head. I was way out of my depth when you got part of the way through.

• 1635

I guess what we're looking for as we move into this kind of legislation and are dealing with this kind of technology is whether or not it is safe, whether it is safe as far as the privacy requirements are concerned, whether the information contained within the samples can be used for other than identification and whether there's any possibility of abuse if the data bank security is breached by someone gaining access to the samples or to the profile of the samples themselves. If you have any comments or assurances you could give us with regard to some of those questions as we look at Bill C-3, I think that certainly I and other members of the committee would appreciate your views on this.

Dr. Ron Fourney: The first thing I'd like to state, from a scientific point of view, is that the information we're collecting is a series of two-digit numbers, and these numbers when they add up essentially identify an anonymous piece of DNA. This has been very carefully chosen so that it doesn't code for any other portions of human disease or mental traits or physical traits. In other words, even if a person received or had these two-digit bar codes across the bottom that I showed you in one of the gels, it would only describe an anonymous piece of DNA. It has no coding value whatsoever for a physical or a mental disease.

Secondly, from the legislation point of view, I know there are going to be strict rules and regulations and penalties governing the misuse of such data, and I would think that's a particular deterrent for anyone who would try to misuse this data, but once again it's an anonymous piece of DNA. I'm not sure how it could be used other than a human identification approach.

Mr. Jack Ramsay: Are you aware of any other DNA data banks?

Dr. Ron Fourney: Yes.

Mr. Jack Ramsay: Of course we all give samples one way or another. I imagine there will come a time when we will not need the root of a hair, so we'll be able to pick up samples at the barber shop and so on. But we give blood samples. I give a blood sample every time I have a medical check-up, that kind of thing. I understand that at birth a sample is taken. I don't know what's done with these samples, whether they're placed in a bank or not. Could you advise us on that?

Dr. Ron Fourney: Actually, it was a question I asked yesterday to some of the clinical geneticists working at the Hospital for Sick Children, and they've advised me they have rules governing their misuse and certainly what they can do from samples that are submitted for disease analysis. So apart from the fact they have collected samples for their own genetic determination, I believe through their ethics board they can get permission to use these—if they strip all the identifiers off—to do research in the sense of trying to identify the linkage of a particular trait through a family. This is the way that myotonic dystrophy genes were cloned, for instance, recently by Dr. Korneluk here in Ottawa. They had permission from the families to look at this trait as it spread through the various family members to try to identify the particular region of DNA that was possibly the signal for the myotonic dystrophy.

So in terms of a clinical diagnostic laboratory this is routinely used, but I think they have strict rules governing how they can use the samples and what they can do with them.

Mr. Jack Ramsay: What about the authority for gathering the samples?

Dr. Ron Fourney: Is that from a clinical diagnostic point of view?

Mr. Jack Ramsay: Yes.

Dr. Ron Fourney: It's my understanding that anyone who submits a sample for testing from a clinical point of view gives written permission for that test to be conducted. So they're in full knowledge of what those samples will be used for.

Mr. Jack Ramsay: So if I gave a blood sample during a medical examination and somewhere along the line I've signed a form, does that allow it to go into a bank?

Dr. Ron Fourney: If you haven't, you will be, because I think the ethics boards from hospitals all across Canada are pursuing this very extensively.

Mr. Jack Ramsay: What happens now at clinics all across the country? What happens with those blood samples?

Dr. Ron Fourney: I would think they're kept in the research laboratory and all their identifications have been stripped.

Mr. Jack Ramsay: That's all I have for this time, Madam Chair.

The Chair: Mr. Discepola.

• 1640

Mr. Nick Discepola (Vaudreuil—Soulanges, Lib.): Thank you, Chair. I apologize for being late, but I was accosted by the representatives from the police association from Toronto, who had me hostage in my own office until I agreed that I had more important things to do here on DNA, and they let me out.

One of the concerns I've had throughout this whole debate is in two areas. One is, we've been told repeatedly that the actual profile, the linkage between the profile and the identifier, could be removed, but that the actual electronic storage—I haven't been able to get a precise answer—of the profile cannot actually be erased.

Dr. Ron Fourney: Were you here during the presentation?

Mr. Nick Discepola: I wasn't here for the first 20 minutes or so.

Dr. Ron Fourney: Okay. I think what I tried to show diagramatically is you'll have an identifier code with a sample coming in, and then the machine will look at the DNA and create a raw data gel image that's unique to the DNA sequencer, and from that we can pursue the interpretation and develop the genetic profile. It has to be linked, the actual identification.

If we cut the bar code or the link either going into the gel file or coming out of the gel file, what we'll have is a gel file with a sample in there that we can see, but we have no idea what it is and we can interpret it no longer. We cannot simply dissect one simple lane out of that gel file, because they're all interconnected. You'll destroy the entire information on that gel.

Mr. Nick Discepola: Why can't you get rid of the whole profile itself as opposed to—

Dr. Ron Fourney: You can destroy all 28 samples if you want, but you would basically destroy the complete gel image.

Mr. Nick Discepola: The legislation calls for removing that link, as opposed to getting rid of the 28 profile samples as you mentioned.

Dr. Ron Fourney: Correct. So what you would do is remove the identifier going into the gel image data collection and you remove it coming out, and after that it would be like looking at a piece of information where we'll see there are 26 lanes but we only have information on 25.

Mr. Nick Discepola: In your opinion, is that sufficient to render that whole profile useless?

Dr. Ron Fourney: Yes.

Mr. Nick Discepola: Okay.

One other debate I have is that the police association wants us to take samples either at the time of arrest or time of charge.

If I listen to your presentation and use the same analogy as fingerprinting, I'm quite convinced that the profile, in itself, is quite useless other than for identification purposes, especially if you've extracted out some of the concerns we had in terms of disease, the chromosomes that identify diseases, and so on.

In your opinion, would you consider that the profile that's generated through the whole DNA analysis is stripped down enough that any concern that individuals might have over privacy issues, over misuse—I'm talking about the profile now, not of the samples—of the profile, maybe in a few years when the public opinion is such that they're more receptive to the use of this DNA and more comfortable with it, we could essentially see an enlargement of the collection of the sample?

Dr. Ron Fourney: I think there are two aspects to your question. First, from a scientific point of view, I cannot see how this two-digit code can be used for any purposes other than human identification. From a non-scientific point of view, I can't really control what will enter or leave the database. That's really up to legislation and regulations.

I would say I'm totally neutral. What I am basically involved with is developing the technology and managing the system for identification purposes.

Mr. Nick Discepola: So in your viewpoint, the two would go hand in hand. In other words, the actual profile should be stored and we should be cautious as to what we store, but for the reasons you have invoked before, the changes in technology and the rapidity with which it changes, we should also store the sample.

Dr. Ron Fourney: I would favour, very much so, that we maintain the initial sample that came in. Of course, in the end, what we are interested in is that two-digit code at the very end. The material in between essentially can be destroyed in terms of the byproducts, the extraction and the amplification.

• 1645

If we ever needed to go back to the DNA analysis in the future—let's say a brand-new technology was developed that was state of the art and would make everything much faster, more efficient and probably less expensive—we could re-analyse those samples. If we don't have the original samples, then we would have to go back and obtain them.

Mr. Nick Discepola: Thank you, Chair.

The Chair: Mr. McKay.

Mr. John McKay (Scarborough East, Lib.): Watching your presentation here, I'm curious as to whether there is any subjectivity left in the analysis.

Dr. Ron Fourney: Subjectivity—

Mr. John McKay: Yes. Is there any room left at all for interpretation?

Dr. Ron Fourney: I've just given you the tip of the iceberg. This is very technical. It involves a lot of different sciences—molecular biology, human population genetics, some aspects of chemistry electrophoresis, etc.

We often get samples from a crime scene point of view that are composed of mixtures. They could be partially degraded. It's very important to have a full grasp and understanding of the limits of the technology so that you do not exceed the capability.

What I'm getting at here is that as the sample becomes more and more degraded it gets smaller and smaller. Eventually some of those little bumps, or the colours you saw, are getting harder and harder to read. We have to have a threshold that we feel beyond which we have no confidence in calling that sample a particular match or a particular fragment. It's at that point that we have full capability through our interpretation to develop the case. It's very important that we understand not only the advantages but also certainly the limitations.

An awful lot of training goes into our scientists to allow them to have a full grasp of what this technology can not and will not do. We routinely have a very extensive two-week in-house training session, with long hours during the course of the day. Those scientists then have to go back and analyse case-file-type gel images we've put together in our laboratory. It takes about six or seven months to do so. We mark all those as proficiency and competency tests. Essentially, an awful lot of rational thought has to go in with an interpretation here.

Mr. John McKay: How do you state that for the purposes of evidence? When you are giving evidence on DNA analysis, how do you state to a court the quality of the sample you received and its limitations?

Dr. Ron Fourney: Basically, if the sample doesn't come up to the quality we expect, the results are inconclusive.

Mr. John McKay: So it's 100% one way or not at all.

Dr. Ron Fourney: Correct.

Mr. John McKay: That's the statement you make to the police.

Dr. Ron Fourney: What we will say is that we may get a partial interpretation of certain profiles. For instance, there have been instances I know of where a sample has been significantly degraded, and we might have only one or two tests we could use. The discrimination is going to be much less on those tests, but it may be enough to provide an investigative lead to individuals.

Mr. John McKay: Do you ever get pushed by the Crown or the police on the quality of the interpretation or the quality of the evidence you can give?

Dr. Ron Fourney: I'm not sure “pushed” is the correct term.

Mr. John McKay: Tested.

The Chair: That would never happen, Mr. McKay.

Mr. John McKay: The issue, Madam Chair, is that there is an appearance of objectivity here that is an overwhelming appearance of objectivity. There is very little you can argue with and use to throw this stuff up, but getting to that point is in some respects a far more problematic course.

I'm wondering whether in fact as a member of that North American group you've established standards to get to that point.

Dr. Ron Fourney: Absolutely. That is one group we're participating in. There's another group, the Technical Working Group in DNA Analysis Methods. We call ourselves “TWIGDAM”. It was developed in 1988, hosted originally by the FBI. We are members of that group. As well, maybe 50 or 60 laboratories participate. We meet two or three times a year.

During those sessions we develop interpretation guidelines and standards of quality assurance, validation, and certainly training standards. They're very extensively developed, and they have to be validated.

• 1650

Through working with other scientists in groups such as those, we come up with cooperative acceptance of not only the interpretation but the limits of the technology.

Mr. John McKay: So if I were a crown or a defence attorney, I would push you in the area of the quality of the sample, the degree of degradation, the means by which the sample was collected, etc.

Dr. Ron Fourney: I would say, as a molecular biologist, the technology is challenging, but we're very satisfied with the results we're going to get.

As many of you are aware from experiences in the past and well-publicized cases, the samples we use to extract the DNA can have varied backgrounds. You can have contamination or what have you. It's also very important to train individuals for collection of such evidence.

In Canada, I think we've done a very good job of collecting the evidence and actually maintaining a very high quality of standard, but the evidence is only as good as what comes into the laboratory.

Mr. John McKay: I have one unrelated question. We've been questioned on the issue of whether the RCMP should run the bank. I'd be interested in your views. Has that ever posed a problem in the past, in terms of questioning the objectivity of the evidence?

Dr. Ron Fourney: Do you mean in terms of being a biased witness?

Mr. John McKay: Yes.

Dr. Ron Fourney: It's certainly a legal ploy that people would challenge us when we're on the stand, but at the same time it's completely understood that a forensic scientist is basically a witness for the court and is completely unbiased for either side. We've testified on behalf of the defence as well as the prosecution.

My first goal is to be an ethical scientist working with the technology and explaining to the court what it can or cannot mean.

Mr. John McKay: Thank you.

The Chair: Mr. Ramsay.

Mr. Jack Ramsay: The conclusive nature of DNA evidence is overwhelming. I sometimes watch that show Law and Order—law and the courts, I guess it is—and I can foresee the possibility of obtaining a sample of your DNA and placing it in the body of the victim at the murder scene.

If that did happen and you were given a sample of sperm found in the body of a murdered victim, which had been planted, the conclusiveness of that evidence is so significant that it raises a concern. We have had two people exonerated from guilt because of DNA alone. There's an opposite side to that coin, in that the conclusive nature of DNA evidence could convict innocent people under certain circumstances.

Do you have concerns in that area?

Dr. Ron Fourney: The first thing I'd like to point out is that DNA does not convict anybody. It is part of the evidence that's provided to the courts, and usually a group of people has to make that decision.

The second aspect is if you found semen in the vaginal cavity of a victim, it might be difficult to explain how it could get there, other than by the means most people would draw as a conclusion.

Third, forensic DNA analysis is just part of the entire puzzle. It is a highly probative tool that allows us tremendous power in discriminating between individuals. But a typical case is built up of very many elements, and DNA could be only one of them. You may have an arson investigation. You could have fingerprints. You could have firearms. You could have other evidence that's outside of the realm of science. This is only part of the puzzle.

• 1655

When I testified in the Legere trial, for instance, there were 42 expert witnesses. Quite a few of them were DNA experts, but I believe that overall there were far more witnesses for other elements of the case, for elements other than the forensic science or the DNA. It was the entire puzzle that was put together that formed the basis for the decision made by the jury.

The other thing I'd like to point out is that there have been instances that I am aware of..one FBI case in particular was about the sexual assault of a young female. The semen stain found on her panties failed to match the accused's DNA. This information was provided in a court of law, and an FBI agent testified on behalf of this defence where the accused's DNA was excluded. In that particular case, they had an overwhelming amount of other evidence—apparently the eyewitness account was excellent—such that the jury found the person guilty of the crime even though the semen found at the scene and the DNA did not match.

So I would say that the entire case has to be reviewed with all the different components, and although we are very excited about DNA technology, it is just one of our forensic tools in the arsenal.

Mr. Jack Ramsay: I have one final question. I don't know whether this is fair or not, but were you able to establish the quality of the DNA evidence in the O.J. Simpson case?

Dr. Ron Fourney: First of all, I didn't have anything to do with the O.J. Simpson case.

Some hon. members: Oh, oh.

Dr. Ron Fourney: It's my understanding, from knowing the individuals involved on both sides as well as the prosecution and the defence, that the DNA itself was not the problem in the O.J. Simpson case. The evidence looked extremely good. Robin Cotton from Cellmark Diagnostics confirmed the DNA conclusions of the California Department of Justice laboratory at Berkeley. The concern in the O.J. Simpson case, from my reading of the accounts, was about aspects in regard to the nature of the collection.

Mr. Jack Ramsay: Thank you.

The Chair: Thank you, Mr. Ramsay. Mr. Maloney.

Mr. John Maloney (Erie—Lincoln, Lib.): What does it cost to analyse a DNA sample?

Dr. Ron Fourney: The current cost for an RCMP DNA case is about $4,500. That's for an entire case. What's hidden in that cost—I shouldn't say hidden—is that an awful lot of quality assurance goes into a particular case. We run positive and negative controls. We will probably re-examine samples that were problematic at first because of extraction, for instance, samples that weren't quite clean enough. And in the separation, I showed you the male and female components.

There are a lot of other things built into the analysis, but from start to finish, including the labour costs and the training of the person—I think even included in that analysis was how much it cost to heat the building, for instance—it's around $4,500 for a complete case. The average case has about six samples of questionable or case-related— and then there are additional samples that are controls we have to use for quality assurance and validation purposes.

Mr. John Maloney: But with this proposed legislation—or perhaps amendments might be put forward—there'll be a lot of DNA samples taken. You can now analyse in a matter of days what used to take weeks and perhaps months. Will this cost factor reduce?

Dr. Ron Fourney: Let me point out something. Don't confuse an operational case with what it will cost to run the database. Remember that in an operational case these are unknown standards that have come in, and they will involve some kind of identification, maybe some presumptive tests. For example, is it blood? Or is it urine? Then they will go through some specialized extraction. The blood taken off a wallboard will probably be extracted slightly differently than the blood on blue denim will, for instance. There are a lot of speciality aspects to an operational case.

• 1700

When the database gets established, we fully intend to use a control standard way of collecting. I showed you some bloodstain cards. We can put blood on that card, do a one-millimetre punch off the card, and process for DNA in about 15 or 20 minutes, having it ready for PCR and running it through the entire test. It's estimated that one sample will cost $50 to $60.

These are estimates based on our predictive value in today's terms. They could change as the technology becomes more sensitive and faster, and other aspects involved with technology are progressive. But we're looking at $50 or $60.

On top of that, if you have all these samples collected in the same manner, it's not unlike a diagnostic lab. If you run a diagnostic test once or twice a week, it's fairly expensive, but if you run the same diagnostic test 50 times in the first day you can actually batch-process the samples, use a series of standards and controls that are similar, and relate back to all those samples, which makes it much more efficient and much more cost-effective.

Mr. John Maloney: When you are doing this batch process, is there a probability of human error that would mix one sample up with another?

Dr. Ron Fourney: We sure hope not, but humans, unfortunately, can create errors.

But I would think one of the things we're dealing with is automation. We're very interested in going to robotics workstations to do a lot of the processing. What you find is that the constant analysis of a series of samples, over and over again, is quite tedious. If a machine can do that, then it can do it better than a human, after weeks and weeks.

The Chair: Mr. Ramsay.

Mr. Jack Ramsay: That was an interesting example you gave about the case in the United States where the samples off the clothing of the rape victim did not match the DNA of the accused. Are you aware of whether or not in that case it was established that whoever raped that woman and murdered her— that it was his semen on her clothing? Was that established?

Dr. Ron Fourney: In the case I spoke of, the victim lived. She eyewitness-identified the accused. His DNA did not match the semen. I'm aware of cases in the past where for reasons other than possibly the actual circumstances surrounding the case the victim may not have wished to disclose where that semen may have come from.

Mr. Jack Ramsay: So the answer to the question is no.

Dr. Ron Fourney: It's my understanding in that case it did not match the husband and it did not match the accused.

Mr. Jack Ramsay: But that's not the question I asked. You may not be able to answer the question. Was it established that the semen on the clothing came from her attacker?

Dr. Ron Fourney: They would not know that. The circumstances of the case, I believe, were that a women was raped in her car. She immediately went to the police station and they took all her clothing and analysed it. The semen found on her panties would have been shortly after the crime had been committed.

Mr. Jack Ramsay: We have had a witness here who I think has an awful lot of integrity and who pointed out that eyewitness testimony is one of the most common frailties in evidence.

Mr. Ron Fourney: Correct.

Mr. Jack Ramsay: Okay, that's fine. Thank you.

The Chair: I wanted to ask a couple of questions about quality control from the point of view of external audits or external investigations of quality control in the lab. We had Dr. Young here from the forensic sciences centre in Toronto. Of course they have had some controversies, not about DNA testing but about other issues of quality control. He talked about the importance of having some kind of quality control body or association which could come in and do an internal audit, from outside, independently, and grade the lab— and to be standards to operate by. Is that something your lab is doing now, or is that something that's realistic?

Dr. Ron Fourney: I think it's realistic. The RCMP is pursuing accreditation through, I believe, the Standards Council of Canada, and the Centre for Forensic Science has chosen ASCLAD, the American Society of Crime Lab Directors, a very good agency.

• 1705

The reason we are looking at the Standards Council of Canada is that they're pursuing a world international standard for quality assurance called ISO 9001, and that's primarily what we want to achieve. Also, I believe that ASCLAD is moving towards a 9001 standard. So in the end, if a laboratory is accredited by a 9001 body, they'll essentially have the equivalent accreditation. So it's very important and it's something we certainly support.

The Chair: Okay.

When Milgaard finally had his DNA testing done, I believe it was done in Scotland or England. Why was that? Why didn't you do it? Why wasn't it done here?

Dr. Ron Fourney: Milgaard's DNA was analysed in Wetherby by the Forensic Science Service. What I can tell you is I'm quite familiar with the case. I was a consultant for the Department of Justice. At the time the decision was made to perform DNA analysis with the STR technology, there were not very many labs that actually pursued it. We were one of the ones that could have done the analysis, and it's my understanding that there were things weighing in other than the science. There might have been aspects of consideration by the Milgaard interests who wished to pursue it elsewhere.

The other thing that should be realized too is that some of the technology that was originally considered to be important for analysis was an older technology, and it's been supplanted by the new STR technology. I think they made a very good decision in using a laboratory that has very much the same protocols and procedures we have. In fact two of the test systems, D-21S11 and FGA, that were used in the analysis at the Forensic Science Service were actually developed and pioneered by our laboratory.

The Chair: So it's not a case of us not having the technology in the RCMP lab here; it was due to other considerations?

Dr. Ron Fourney: Correct.

The Chair: Mr. Ramsay was talking about his television-watching habits. I watch Nova, and one of my favourite Nova shows is the one where they test the woman who claimed to be Anastasia. I really liked that one.

What concerns me about that is we were talking about biological samples from people—blood samples and that sort of thing that are kicking around in hospital labs—and the reason they could prove that she wasn't Anastasia was that they had blood from the Duke of Edinburgh and they had a tissue sample from her from some surgery she had had. Next week, if I kill someone, I'm a little bit concerned, because there's probably a chunk of me sitting in a lab somewhere, because I've had surgery as well. So how do we control that, or do we? Or is this outside your area of expertise? When you're testing my DNA from that crime site, are you going to be looking for that chunk?

Dr. Ron Fourney: What I can tell you is it comes back to the basic, fundamental aspect of forensic science, which is continuity of the samples. Do we have full understanding of where the sample was collected, how it was collected, and the chain of custody linking all the way through? I would say if someone deposited a sample to me and they didn't really know exactly where it came from, it would not be tested. It would have no implication whatsoever on that case.

You do raise an interesting question in the fact that the analysis for that particular case was done, I believe, by the Armed Forces Institute of Pathology, Mitch Holland's group down in Bethesda, and they used a technology called mitochondrial DNA analysis.

The Chair: Right.

Dr. Ron Fourney: That's a little bit different from what we're using today in our own laboratories. We have the capabilities of mitochondrial, but we have not pursued it for forensics, because the major reason mitochondrial DNA is of benefit in a case like this is it's maternally inherited. It's a different type of DNA from what we use, and essentially the DNA is passed down through your mother, your grandmother, etc., and that's how they could link it back to the Duke of Edinburgh.

The Chair: Yes, it was a cool show.

Are there any other questions?

Mr. John McKay: When is Nova on?

The Chair: Nova is on twice a week in Windsor on channel 56 from Detroit.

All right. Thank you very much, Mr. Fourney. That was very interesting.

Thank you, colleagues. We'll rise now.