Call Dr. John Gerdes.
JOHN GERDES, called as a witness on behalf of the Defendants, was duly sworn and testified as follows:
You do solemnly swear that the testimony you may give in the cause now pending before this court shall be the truth, the whole truth and nothing but the truth, so help you God?
(BY MR. BLASIER) Let me show you what's been marked as 2263.
Is that an accurate copy or current copy of your curriculum vitae?
And can you please briefly describe your educational background in the field of molecular biology?
Yes. I received a Bachelor of Science degree, University of Wyoming, in microbiology, and proceeded to U.C.L.A., University of California at Los Angeles, where I did a Ph.D. thesis on microbial genetics and also microbial pathogenesis. And microbial genetics is a predecessor field, if you will, of molecular biology.
And then I did a post-doctoral fellowship, again at U.C.L.A., investigating the molecular biology of herpes simplex virus and investigating the mechanism of causing latent infection in animal models.
And subsequent to that, I went to University of Colorado in Denver, Colorado, Health Science Center, the medical school, where I was an assistant professor for four years.
Primarily virology, which is a study of viruses, and some molecular biology for medical students.
Then I went to Hawaii for about five years, where I taught at a community college, and also was a quality control microbiologist at this time.
And I returned to Denver, spent one year as director of a laboratory investigating multiple sclerosis and investigating the detection of viruses in brain tissue from autopsies of multiple sclerosis patients and trying to find out if we could determine or identify a virus as a cause of multiple sclerosis was our goal.
And then I joined -- or was hired in my current position at Immunological Associates in 1988.
And the study of viruses and what you described in your educational background, does that deal with issues relating to DNA?
Well, a virus is simply a piece of DNA that's wrapped in a protein coat. Either DNA or RNA. RNA is a second type of nucleic acid. In investigating viruses, you're also investigating nucleic acid.
I actually have three titles.
As clinical director, I'm responsible for the testing that involves detection of infectious agents and any clinical chemistry types of tests we do related to -- specifically what we are is a transplant laboratory, so organ transplants.
And what we do is we investigate infectious -- obviously, we're interested in whether or not -- if you have a potential donor, it's our responsibility to ensure that not only do you match the immune system of that particular donor organ with the recipient, but also ensure that it's not infectious.
So you need to look for infectious agents to ensure that you don't inadvertently pass an infectious agent into the recipient and therefore give them an infection.
So that's my clinical title.
And I'm also the director of DNA parentage analysis. Our lab does paternity testing or parental testing. It is for the purpose of the legal -- it's an interface between science and the legal or law and involves any situation in which there's a question as to the -- the parent of a child. So that would -- can determine whether a father is indeed the biological father or not. That kind of question, that involves DNA testing. We use both RFLP type of testing as well as PCR base testing now for that purpose.
And my third title is director of research and development, and it's my responsibility to identify testing that's new, testing that can be helpful for us to introduce into the laboratory, validate it for that -- whatever purpose that new testing would have.
And in addition, I've received a grant to investigate a new strategy of doing DNA testing in such a way that it could preclude the possibility of contamination.
In the work that you've done in your laboratory, have you become familiar with the DQ Alpha HLA region of the DNA, the DQ Alpha gene?
Yes. The HLA region actually involves several different gene loci, and DQ Alpha is one of them. And we're very familiar with that because of the fact that the other thing our laboratory does -- and I actually was involved in setting this up, but we now have another Ph.D. that has this department of the lab.
But what we do is molecular DNA dot blot base testing for the purpose of matching the organ or matching a bone marrow transplant for a recipient.
That's by DNA base testing of a dot blot format and that involves matching in the HLA region because the HLA region is that area of DNA that's involved in the immune recognition. So that's the critical area that you need to match between the donor and the recipient to ensure that the -- there's no immune rejection of the grafted bone marrow.
It's very critical that you match that very precisely so the individual doesn't reject their transplant.
Now, are you familiar with the DQ Alpha testing kit that is used for forensic applications, or analysis of the DQ Alpha gene?
Now -- incidentally, with respect to the DQ Alpha gene, how many alleles are there in the DQ Alpha gene?
Well, the kit detects 6 alleles. In fact, there are -- there are more alleles than that; perhaps twice that many. And there are new ones discovered all the time and -- but the kit was designed to detect 6, but in fact there could be as many as 12 to 14 now.
Now, in your work with transplants, in terms of matching the donor to a recipient, you use the DQ Alpha gene system, do you not?
Is the 6 allele system that is in the kit used in forensics, adequate for transplant technology?
No. We use a direct dot blot method that detects more of the -- all of the alleles, in fact.
Well, in -- in a situation where you're doing a bone marrow transplant, you want to match as precisely as you possibly can. And so we use the highest -- the test with the greatest discrimination in order to -- to do that matching.
Now, let me ask you about professional societies that you belong to in areas relevant to your testimony.
Tell us what those are?
Well, I belong to the American Society for Histocompatibility of Immunogenetics, which is ASHI for short. That's an organization that's responsible for accrediting laboratories for the purpose of doing HLA diagnostics, specifically HLA diagnostics relating to transplant work.
I'm a member of the American Association of Blood Banks, AABB for short. AABB is the organization involved in accreditation and inspection of laboratories that do DNA testing for the purpose of parentage testing.
I belong to the American Society of Microbiology Clinical Ligand, L-i-g-a-n-d, as an associate. A ligand is just a target. It's another word for that. It's a clinical target. This organization is involved in investigating that, or basically it's a collection of individuals whose job titles involve clinical testing.
And I believe --
Yes. That's a society for virologists and it's a collection of directors of clinical virology laboratories.
During some point in your career, did you become interested in the forensic issues of different technologies?
I believe that was back in 1990. At the time, we were doing DNA testing for parentage testing and had an advertisement that was in the yellow pages for that purpose.
And there was a local murder case and I -- the story is that the attorney just looked me up in the yellow pages and called me and said would you look at this testing.
And I did.
And I had some concerns and expressed those concerns about that particular case.
And then subsequent to that, I was invited to give a talk at a Defense Attorneys Conference in Colorado, and discussed specifically RFLP in that conference. But I met several individuals; Simon Ford and Bill Thompson. From that time on I started to get referrals. After discussing my concerns with them, I started to get referrals of cases, forensic cases.
In your -- your work preparing forensic cases, have you been called upon to visit forensic labs?
Well, any laboratory has a written document that is called their standing or standard operating protocol. And what that is is it's just a written -- the written directions, if you will, or instructions that the technicians need to follow. And so by reviewing those instructions, one can determine the policies of the laboratory without actually going to the laboratory.
And as far as the labs that you visited, is one of those labs a Los Angeles Police Department SID lab?
Well, my usual procedure in looking at a laboratory is I ask them first to give me a tour of the lab, and what I look for in a tour of the lab is if, for instance, they're taking PCR base testing, I'm looking for the organization of the laboratory in terms of how the laboratory deals with common problems in PCR testing, mainly the fact that you can have contamination.
So we look for how they deal with that, and what precautions they take for that, and whether their laboratory -- what their laboratory -- the appearance of the laboratory in terms of all of the aspects that might have to do with that.
After going through the laboratory tour -- and during the laboratory tour I usually just ask the director to tell me, well, assume that a specimen just came in the door, it's from a case, how would you handle that specimen, and walk me through the physical locations of that evidence of that specimen, that reference specimen, and how would you do that.
After that, I usually ask to see records of all of the testing that has been done, that tracks the test that's of interest or that -- the case that I'm looking at.
And the reason for that is to get an impression of how good the laboratory is in term of not having incidents of contamination.
So I can look at the control strips that are run around the same time as the case that I've been hired to look at.
And I get some impression about how clean the laboratory is, how well their protocols -- even though they have -- they may have a very good written protocol on paper, but that gives you a window of looking at how well those protocols are actually followed, or how effective they are at actually preventing contamination. You can look at a window of time around the testing as far as their control strips, look at a lot of those control strips and see if there's any problems in the lab.
Then I ask to look at their proficiency testing. Proficiency testing is just a -- a mechanism of testing the laboratory to see how effective their -- their typings are, and there -- there's a number of organizations such as the College of American Pathology that send out blind specimens that the laboratory doesn't know the results of, and they just type those and then send back the results. So I look at those testing events to see how well they did on those tests.
Are all those steps important in assessing the reliability of test results in a particular case?
Well, we're accredited by a number of organizations. There's actually four of them, and I've mentioned them already.
The American Association of Blood Banks accredits our parentage testing by DNA.
The American Society of Histocompatibility and Immunogenetics, or ASHI, accredits our HLA testing, and that includes HLA DNA testing.
The National Marrow Donor Program doesn't actually accredit. They're an organization that is responsible for typing specimens for the purpose of bone marrow transplants. They only contract labs that are ASHI accredited so they depend on ASHI accreditation, but they -- they perform what are called blind proficiency for their contract labs.
What that means is we have a certain number of individuals we have to do the typing on every week, and a percentage of those, about 10 percent of those are actually controls that -- we don't even know which ones are the controls. So the specimens that come in are all just given numbers -- accession numbers. That just means each sample that comes in is given the next number in sequence. So we don't know who the individuals are, we don't know which ones of those specimens are controls. We run those tests, and then every week we're graded in terms of how well we've done on the control samples.
So they're a proficiency organization as well as the regular industry organization.
And then there's the CLIA. This is the Department of Health and Human Services. The federal law for clinical diagnostic testing is called CLIA, Clinical Laboratory Improvement Act of '88.
That's a federal law.
According to that federal law, inspectors, I think it's every two years, inspect your laboratory and accredit those laboratories for the purpose of doing any specimen that will be used in human diagnosis. And we're accredited by them as well.
Is it accurate to say that to maintain your accreditation, you have to subject yourself to the blind proficiency testing?
That's true in the case of NMDP, the National Marrow Donor Program. The CLIA regulations at the present time, they all -- all of these accrediting agencies require proficiency testing, but the distinction is that only NMDP requires them to be blind in terms of not knowing which ones are the test specimens.
All of the others involved, on a monthly basis or on a quarterly basis, specimens that come into the laboratory that need to be tested, but we know that they're quality control specimens that are coming in the door.
But we have to do proficiency for all of those organizations.
And is it accurate that the Los Angeles Police Department Scientific Investigation Division lab is not accredited by anyone?
Now, approximately how many cases have you worked on -- how many forensic cases have you worked on since you became interested in approximately 1990?
There are 35 that I've actually testified in. There are perhaps another half dozen where I've consulted but I did end up testifying. And there's perhaps two or three cases right now that I'm involved with.
And you spent a considerable amount of time reviewing materials for the Simpson criminal case, did you not?
Now, you already indicated you reviewed all the protocols and did lab visits for all three labs.
Have you reviewed the collection procedures used in this case by the Los Angeles Police Department?
And have you reviewed testimony by the various people who collected the evidence in this case?
How about in terms of processing of those samples once they're collected, what have you reviewed with respect to that?
That would be Collin Yamauchi as well as the other DNA individuals, Robin Cotton and Gary Sims.
Have you reviewed all of the PCR testing results obtained by all three of those labs in this case?
Have you also reviewed a video that was prepared by the Los Angeles Police Department to show how they used -- how they collected evidence in this case?
We'll get to that in a minute.
Do you have an opinion with respect to the risk of error in the results in this case due to contamination at the Los Angeles Police Department.
(BY MR. BLASIER) As a molecular biologist, Dr. Gerdes, and a DNA laboratory director, do you have an opinion about the specimen handling and sample methods used by the LAPD personnel in this case?
I believe there's an unacceptable risk or chance in the way those samples were handled of there being cross-contamination.
KEY QUOTEDo you have an opinion with respect to the reliability of the PCR testing results obtained in this case by the LA Police Department?
Objection, Your Honor, same objection as before. This is trying to incorporate the outlawed testimony.
Mr. Blasier, I know you understand my ruling and I trust you are being guided by it.
I asked you a simple question and you're answering yes. So you may proceed.
KEY QUOTEDo you have an opinion -- a professional opinion, as a laboratory director, and given the evaluation of the results in this case that you have done with respect to the reliability, to a reasonable degree of scientific certainty, of the PCR results obtained in this case by the Los Angeles Police Department?
Again, I feel that the sample collection, manipulation and PCR setup protocols are such that there's an unacceptable risk of cross-contamination.
Do you have an opinion, same question, with respect to the PCR results obtained in this case by the Department of Justice?
Well, I believe that the -- the results -- the Department of Justice results are reliable. However, with the qualification that whatever -- their typing went through the LAPD first, and so if there were any errors made in terms of cross-contamination at the LAPD, the Department of Justice would -- Justice would simply repeat those errors. The other qualification is that the Department of Justice has an artifact that's frequently found in their testing strips which is a 1.3 allele, which makes the interpretation of results in a number of specimens suspect.
Do you have an opinion with -- we'll talk about that in specific.
Do you have an opinion, same question, with respect to the PCR results obtained by Cellmark Diagnostics in this case?
It's similar to the DOJ, and that is that I believe the Cellmark DQ Alpha system is -- poly marker system, excuse me, or both, is run reliably at Cellmark laboratories. However, they are going to again retype any errors or just amplify those errors that occurred prior to the samples coming to them.
And again, there are some occasions of contamination in a number of the specimens -- specific specimens that were run at Cellmark.
Okay.
Now, we've had some testimony about the comparison of the use of DNA technology in the clinical setting versus the forensic setting. You have some views about that, correct?
Is there -- there is a significant difference between the type of DNA testing that you do, or that's done in a clinical setting, and in the forensic setting with respect to the -- the type of samples that are used?
Well, the samples -- in a clinical setting we receive samples that are sterile samples, that have been collected by a trained nurse or individual who knows how to follow what's called an aseptic technique, which is a manner in which to collect a specimen with sterile protocols so that it's a clean specimen.
Well, it's extremely important especially if you're doing something that involves identity testing.
Because of the fact that's in a clinical sample, then we can guarantee that that came from a single individual.
And so if there's any spurious results, meaning that we have some signals on our typing strips that shouldn't be there, then we know that there's something wrong, and we need to have that specimen redrawn, and we can retype it and repeat it.
Now, in a forensic setting, the samples are -- are dirty, they're not aseptically collected, they can't be because of the nature of -- just crime scenes, and therefore when we have those unusual results, it's very difficult to determine with any certainty whether or not there was a problem with the way the test was run, or there's contamination, that is, human DNA got in there that shouldn't be in there.
That's what the definition of contamination is; there's foreign DNA, somebody else's DNA somehow got into that sample at some point, or that there's some artifact that happened in the way that the testing is done. You can't really distinguish between those possibilities.
So basically -- and the other possibility is that it's a true mixture that, yes, at the crime scene there was more than one person there.
And obviously, it's very important to be able to distinguish between whether an allele that you observed there is a second individual contributing to that sample. It's very important to understand, or be able to say with certainty that that additional allele that is there is due to in fact somebody at the crime scene as opposed to an artifact that happened in the lab.
Now, an artifact is just -- it's a defect. And these dot blot based testing systems where we can sometimes get weak signals on these strips that are due to an artifact, due to a defect in the way that the testing is set up, and so it's not a true result, it's a mistaken result.
So it's very important to be able to distinguish between whether there's a true result there or a mistaken result, obviously.
And in the clinical setting we can because we know that the specimen we get came from a single individual, and we know that anything there that's unexpected, as far as additional signals or alleles indicating an additional person, means that there's something that has gone wrong in the testing, and so you can redo the testing.
Now, is there a difference between forensic applications of this technology and clinical applications with respect to the size of samples that you have to work with?
Well, in a clinical -- in the vast majority of clinical testing you have a generous sample size, a blood specimen that's asteriley collected blood specimen, you can ask them to collect whatever amount you need for the particular test. And usually, for these PCR-based tests you don't need a lot. Certainly, you can request a blood tube which contains around 6 to 8 cc's of blood. So that's plenty. That's a lot of -- of blood to be able to work with that's clean blood.
In a forensic setting there's frequently very minuscule size samples; a small blood drop that's found at the crime scene, a hair that's found at the crime scene, the very small flecks of blood, and it's much more difficult to get a correct typing result when you're limited in the amount of material you have to start with.
Incidentally, let me ask you, can you give us an estimate of the number of nanograms of DNA in one and a half cc's of blood?
And I don't mean to put you on the spot.
Well, let's put it this way: A drop of blood would have about a thousand nanograms, and a drop would be about 50 microliters.
And is it accurate that in the samples in this case, in the Simpson criminal case, in the -- in the civil case, the amount of DNA involved in many of these samples that were subjected to PCR testing is extremely small?
Now, is there a difference in terms of reliability of results between the clinical setting and the forensic setting with respect to how often samples have to be handled?
Yes. In a medical setting, again, we get a sample which was aseptically collected, it's transported to the laboratory, and then a specimen is withdrawn directly from the blood tube, and the DNA is extracted directly from there.
In forensic -- in a typical forensic case, there are swatches and multiple swatches that are taken of stains of blood, for instance, and then those are manipulated in terms of they have to be dried, they have to be packaged at the crime scene, they have to be dried so that you -- that you don't have degradation, they have to be transferred into the bindles.
And so there are multiple manipulations where they're like shuffling these little swatches around.
Every -- each time you manipulate a sample, it; number one, increases the possibility of an error occurring; number two, increases the possibility of the introduction of foreign DNA or contaminants into those particular samples.
Is the issue of error rates of a particular laboratory important in assessing the reliability of results of a particular lab?
Yes, I believe error rates are very relevant. And it has to do with the fact that in a typical identity test DNA figures will be quoted in terms of the likelihood of a match being in a certain population, and then -- and they're very high numbers.
And yet if you -- if you look at the testing of the usual laboratory in a clinical setting, for example, where there have been control studies that have been done, the usual type of error rate in a clinical lab is somewhere around 1 percent. It has to do with the fact that there's human error, and that's about the rate at which people make mistakes. And that's in the best clinical labs.
And so error rate generally tends to be much higher than the gene frequency estimate. And so in my mind, it just makes sense that you would look at both of those and say that, well, if the gene frequency is 1 in 10 billion or whatever, but the lab makes a mistake in 1 in 100, you need to know both of those, all of that information, in terms of how much weight to put on that evidence.
Now, the number 1 percent, does that come from evaluation of proficiency testing that's required to be done?
Now, with respect to laboratory standards -- I think you already talked about this.
Basically, the clinical labs have very stringent standards about sample handling and how they process a sample from start to finish, correct?
In your experience, do the procedures used by the police department SID division in this case, required the same -- require the same kind of rigorous controls?
Now, in the DNA testing that you do, and that's done in the clinical setting, do you ever get into the statistical controversy in terms of trying to predict percentages of populations that might fit a particular pattern?
We do statistics in association with parentage testing which is not identical but similar. I'm aware of the statistical issues.
Correct. In a typical clinical setting, an HLA typing, where we do the typing for the purpose of matching donor and a recipient, there's no statistics involved at all.
So you really don't -- it's not a concern, it's not something that you're -- that's an issue because, you know, you have both individuals you try to match, they're known individuals, and you're just asking the question, does the sample from individual 1 and individual 2, do they match. And you're not concerned about, well, how often would that happen at random in the population. All you're really concerned about is how close the match is between those two specific individuals. So statistics aren't even involved.
Does the affect of environmental conditions on a forensic stain, does that affect the ability, to or the reliability, of results that you can ultimately get from DNA testing?
In my experience, yes, because it's -- again, it's not when you -- take a specimen that's been taken off of carpet; there can be carpet dye or shampoo or anything in that carpet. Or off of a dirty sidewalk; dirt has been shown to interfere with the extraction and the ability to type DNA. So -- and in a forensic setting there's an infinite number of different substrates -- those are called substrates, different areas where you can find a specimen.
And all of those can affect DNA typing.
In the clinical setting are you ever put in a situation where you collect a stain off of a ground somewhere where you don't know how long it's been there or what conditions it's been subjected to?
Now, let's talk about contamination in DNA testing.
Incidentally, is contamination a big issue in the clinical setting as well as the forensic setting?
Well, because of the fact that this is an extremely sensitive technique. It literally has the potential of detecting a single copy of the gene you're looking at.
And so with that extreme sensitivity, that means that it's very easy if you're working at very -- at sensitivity levels where you're demanding that the test detect very low numbers of copy, it's very easy for a foreign DNA or extraneous DNA to be introduced into that sample and appear as a mixture, which would be a false mixture if it's accidentally introduced, and that's called contamination.
And with this technique, it's so sensitive that it literally could wipe this bench and get somebody's DNA, whoever sneezed here last, probably, you could get their DNA type off this bench. It's that sensitive.
So it's sensitive enough that you begin to pick up DNA that's sort of a background DNA in the environment.
And in a laboratory that creates problems because, obviously, you're trying to only get a DNA type for the specimen you're interested in. And the fact that it is so sensitive to that extraneous DNA makes it extremely difficult to control that.
Does that sensitivity come from the fact that small amounts of DNA in the PCR process are multiplied or amplified to many copies?
That's right. The PCR, polymer chain reaction, it basically starts with a very small amount of material and then copies that millions and billions of times so that you have many, many copies of what you started with.
Can you just tell us approximately how big -- What's another size of 20 nanograms of DNA?
20.
20 nanograms would probably be somewhere around five millimeters to a centimeter, something like that I would think.
(BY MR. BLASIER) You remember there being discussions in this criminal case about whether DNA could fly or not?
(BY MR. BLASIER) Can you describe some of the ways in which DNA gets moved from one place to another without you being aware of it.
Yes.
One of the most subversive ways, and especially in forensic cases, one that is of greatest concern is what's called cross-contamination.
There's definitions for different -- so basically three different ways you can contaminate:
The first one is cross-contamination. That has to do with the fact that if you touch an item that has DNA on it, such as a blood stain that's a large blood stain, you can get some of that on your glove, and because we just said that it's very, very small, essentially you wouldn't even notice or see it on your glove, you can still have plenty of DNA there to transfer.
What can happen is you can get that DNA on your glove, it doesn't have to be a glove, it can be the instrument that you pick that up with, or whatever, and then if the next item you handle happens to be an item that's very -- is an evidence item that has very little DNA, where it's extremely degraded, then you can transfer the DNA from the larger amount to the second one that has the smaller amount.
And when you type that in the laboratory, what can happen is now you'll see a match between the two items, but the match was created by the accidental transfer that occurred onto your glove.
Now, that can happen -- as I said, it's extremely subversive. It can be as simple as perhaps you collect an item and you don't change your gloves, and pick up a pen. Now it's on your pen. You may change your gloves but then you pick up the same pen. Now it's on your new glove.
And you can see how easy it is to really transfer DNA around without being aware of it.
And you can think back and you really don't remember how you possibly could have touched one item to another.
And there are techniques that allow you to avoid that. And it's basically developing a second sense of the -- those kind of -- of motions and movements that you don't even think about. And in microbiology it's called aseptic technique; you learn -- if you work with an organism that's a dangerous organism that can infect, you learn not to do that kind of thing. Now, even if -- even in the best laboratories it can still happen that you have an inadvertent transfer from one item to another; cross-contamination. It can happen at any stage, at any time when samples are manipulated.
And could anything happen if you wear glasses, and happen to adjust your glasses as you're doing your work?
Yes. It's those kinds of subconscious movements. You're not aware of it. You might -- I have -- everybody who wears glasses has a constant habit of just adjusting them. They fall down on your nose.
But if you're doing this kind of technology, before you adjust your glasses, you'd have to change those gloves because you put it on your glasses, then you're not aware of it, next time you touch your glasses, it's on your glove, you transferred, now you're transferring to the next item.
So there are an infinite number of ways, when you really think about it, how you can transfer things without being aware of it, from one item to the other.
Now, one of the things that you evaluated in connection with the results in this case was the procedures used by the Department of Justice to protect just against that kind of contamination?
Yes. I mean he -- I think Gary goes-- he does an excellent job of trying to avoid this kind of problem. And he goes -- he's tried to think of everything.
So, for instance, if you were to have notes -- and as you know everything has to be documented. And in these kinds of cases there's -- every time you're handling a specimen you're going to be recording that fact, and if your DNA happens to contaminate those notes, the notes following the specimen, so now it's on the notes. And you handle the notes the next time, now it's going to be on the -- even if you got fresh blood, you transfer it off the notes onto the next item, or at least that's potentially possible.
He actually takes the notes and he exposes them to UV which kills DNA between each manipulation.
That's how paranoid you get at the possibility of this kind of transfer.
Are there other procedures with respect to cleaning the work area between handling samples?
In most forensic laboratories, in between each item you would totally change the paper, preferably bleach the area down. Bleach kills DNA. Bleach the area down and change the paper in between each manipulation of every evidence item.
It makes it an extremely slow, painstaking process, but it's a precaution that most forensic laboratories feel is necessary to avoid this kind of cross-contamination transfer.
Now, is there another way to illustrate how a substance like DNA can move from one place to another without you even knowing it, an analogy with respect to getting a cold?
Yeah. It gets back to your question about can DNA fly.
Well, it doesn't have wings so it can't fly in that sense. But certainly DNA is very analogous to an infectious agent in terms of -- as I said, a virus is DNA wrapped up in a protein coat. And basically, it can introduce -- be introduced into what is called an aerosol.
An aerosol is fine droplets of moisture that can be suspended in the air. And those drops of moisture can then attach to dust, and it can float around in the air for actually quite a while. And that's -- that's really not farfetched because the analogy is catching a cold.
The way we all catch a cold is exactly by that transfer; someone sneezes, I've got droplets in the air and you happen to walk into the room a few minutes later and breathe that in. Now I've transferred the virus from individual 1 to individual 2. It's a very small amount of material that's transferred. When you breath it in, the virus replicates itself, copies itself like the PCR, and it gives you the cold.
So if you think back, how many of you really know the last time you had a cold, can you identify a specific incident or a specific individual you know gave you that cold.
Most of us can't.
You have no idea. You just know I caught a cold somewhere, and along the way you had to have -- to have had DNA transfer into your bodies for you to catch that cold.
So it shows you that this PCR process is every bit as sensitive as an infectious process, and in detecting that transfer.
Let me ask you a little bit more about contamination. You say there are three kinds. The first one was cross-contamination?
Cross-contamination is the transfer of DNA by manipulating an item and then accidentally transferring DNA on your instruments, or whatever, pencil, notes, whatever, to the next item.
And what are the other two kinds of contamination that you can see in these kinds of tests?
The second kind of contamination is called reagent contamination. Reagents are the fluids or liquids that we have to use for the methods of extracting and amplifying this DNA.
So we mix together these -- it's kind of like if you're baking. It's the ingredients list. You mix these ingredients together. That's what allows you to create an environment so that the DNA can copy itself. So you have to add a certain amount of this and a certain amount of that in order for it to -- for this reaction to go.
Those reagents are generally in much larger volumes than you would use just to do a single item.
So, for instance, you might be able to do 100 or a thousand or 10,000 different tests by using these bottles of reagent.
So if you have a bottle that's like the size of, you know, a typical mouthwash bottle type size, then you're only using a drop at a time.
And that means you're going to be going in and out of that bottle a lot of times when you do this DNA testing. And every time you open the lid, you're exposing that bottle to the possibility of DNA falling in there.
And in fact, that does happen.
And when it happens, then if you were to just take the fluid that you need to put in -- the ingredients you need to put in the PCR, and you leave the DNA out, then all of the sudden you start seeing DNA where it wasn't there before, because it actually accidentally got introduced into that bottle, then that's going to be introduced on top of the sample that you're trying to test.
So that's called reagent contamination.
Did you make an observation, in your visit to the LAPD lab, with respect to the chemicals that they were using in their DNA lab in the tests they did in this case?
If he was present during the testing of the evidence in this case, you may examine him.
(BY MR. BLASIER) You're talking about chemicals that were used in the lab in the testing in this case?
Yes. The lots of some of the reagents that were used in this particular case were up to six months old.
Well, because the longer you keep those reagents around, the higher the risk or chance of some of that DNA -- extraneous DNA being introduced, and it being a contaminant.
It depends on the reagent.
Most of them are aloquated (phonetic) so they're used only once. If we're doing an extremely sensitive technique, we use a technique to detect the virus, for instance, that is present in very low copy in transplant patients, and we have everything set up so we only use that once and throw it away. It's disposable. We never go into it twice. Other techniques where we're not working at quite as high sensitivity such as HLA typing, reagents, we probably change every week or perhaps every two weeks.
The third kind of contamination has to do with the fact that in this process of PCR, what you do is you identify a very specific target, piece of DNA, and you copy that millions and billions of times.
If you do that in a laboratory over and over, day in, day out, that's your routine, you're always copying that same -- same piece of DNA, eventually you get a buildup. And those billions of copies are being handled in the laboratory in the same environment where you're trying to find, you know, a few copies.
And so it's very easy to cross-contaminate the previously amplified product and -- to what hasn't been amplified before, and then reamplify it. And that's called amplification carryover, amplification contamination.
When, as I said, I go into the laboratory and look at the physical design of the lab, that's pretty much what I was looking for, this one-way work flow.
It has to do with the fact that in order to try and avoid this contamination with the previously amplified materials into things they're trying to evidence, items or whatever, that have very low copies, if you design the laboratory so that you always have the evidence or the small copy numbers items come in the door, and then as you go through the process and end up with the amplified material, you never have cross-talk between or physical connection between the post amplification area and the preamplification area. Then you can, to a certain degree, avoid this kind of transfer from amplified product contaminating the unamplified materials.
Now, the PCR test that was done by LAPD in this case, what was their procedure with respect to this amplified product and what they did with it?
Well, they amplified -- they do the extraction in one location or building, and then they go to a second building to do their post amplification. That's good because that -- I mean you can't get better separation than that. But --
Correct. They do it at Parker Center, they do the amplification. And at Piper, they do the preamplification. And then they take the amplified product and do their gels back at the original lab.
So they're going not one way, but this direction and then back.
(Continuing.) And even though it's done in a separate room, it still means that amplified product is being carried back to that lab and has a potential of cross-contaminating or contamination by this method.
What sort of controls are used or were used in this case to -- well, first of all, what's a control?
Well, because we're aware that there are problems, there are potential problems in terms of this contamination issue, especially, or, you know, artifacts that may happen in testing, controls can be designed to try and detect those artifacts or those contaminants, and the controls then are used to determine whether or not there's an issue -- that those are issues.
Now, the problem with controls is sometimes, not always, do those controls detect a problem. But if they do detect a problem, then it does tell the laboratory that -- that they have suspicious typing results and they need to repeat it or do something to it. It affects how you interpret that result.
So the kind of controls then is -- we start back at where you start. You start where you manipulate the specimen. So at the crime scene, when you're collecting the specimen, there's a control called a substrate control, which is taken at the same time that you collect the evidence item.
And what that is, is, it's just a blank piece of swatch that is ideally manipulated immediately after manipulating the evidence item that was collected. And that -- that swatch should be a sterile swatch that doesn't have any DNA at all on it. When this goes -- then, from that point on, you have to intersperse one of those control substrate swatches in between every evidence item; so you do an evidence item which has DNA, a control substrate swatch which shouldn't have any DNA, an evidence item that should have DNA, a control substrate swatch that shouldn't have any DNA, and so forth.
Now, you not only have to collect it that way, in order for this control to be effective, you have to be absolutely sure that at every subsequent step that same sequence is followed, and that is you always manipulate the control swatch at the same time you manipulate the evidence swatch, in the same way, with the same tools.
And that hopefully would allow you to pick up this cross-transfer because if there's cross-transfer it should presumably be transferred onto that control.
Yes. As far as the sequence of always handling the swatch, the control swatch, at the same time that you control -- that you handled the evidence swatch, there's evidence that that chain of doing these things in series was not followed at all -- on all occasions.
Well, in this particular case there were two occasions where samples were packaged and mailed to other laboratories; they were mailed to the Department of Justice, and they were mailed to Cellmark.
Now, if this protocol was stringently adhered to and understood that it needed to be stringently adhered to at every stage of manipulation, then when those samples were sent, they should have sent the control swatches because they had -- if you're going to package it, you have to package the control, handle it the same way, and make sure you're following that protocol to make sure there's not transfer.
But, in fact, those specimens -- the evidence items were mailed to the DOJ and Cellmark, and the controls weren't mailed with them. So that says they weren't following this as religiously as they should be as far as handling the control always at the same time as the evidence.
Now, you're familiar with the National Research Counsel Report on Forensic Applications of DNA Technology issued in 1992, are you not?
I believe there's an unacceptable risk or chance in the way those samples were handled of there being cross-contamination.
It literally has the potential of detecting a single copy of the gene you're looking at... it's sensitive enough that you begin to pick up DNA that's sort of a background DNA in the environment.
Most of us can't. You have no idea. You just know I caught a cold somewhere, and along the way you had to have had DNA transfer into your bodies for you to catch that cold.
The Department of Justice results are reliable. However, with the qualification that whatever — their typing went through the LAPD first, and so if there were any errors made in terms of cross-contamination at the LAPD, the Department of Justice would simply repeat those errors.
I asked you a simple question and you're answering yes. So you may proceed.