Dr. Cotton, I'd like to touch on a few of the topics that Mr. Blasier talked about on cross-examination.
First he talked a little bit about the fact that he tried to give you this example of making the DNA big enough to go all the way around the world. Remember that example?
In the whole nucleus, all the chromosomes comes to about 6 billion. That again is an estimate.
And of all that DNA, I think you testified previously that some percentage of it is common to all of us?
So there would be no reason to test any of that DNA that's common to all of us, would there?
So there's no reason to try to test this long strand of DNA, but rather focus your tests on the portion that is unique; is that right?
If you're asking a question about identification, then you only want to bother to test those portions that are different from one person to the next.
And therefore in these RFLP probes, you've been talking about where I forget how many base pairs you said you tested in the RFLP probes, each one of them was about how many?
Well, the average is going to be maybe 8 or 10,000. But even so, even if you added up all the number of base pairs that you're testing, you're still testing a very small amount, even of the part that's different. But you're testing the part that's giving you an enormous amount of information.
And the tests are described, designed and focus on areas where you can gather a lot of information about identity?
Let me give you two examples that are connected, tissue transplantation, if somebody is looking for a kidney or a liver or a heart. Those tests are looking at particular parts of the DNA and those tests are generally done using PCR and not too different from what we're talking about here.
In cases of bone marrow transplants where you have a person who's ill, they're getting the transplant and you have a donor of the transplant, those people may be related because they're related.
It's harder to tell them apart, even with their DNA and RFLP testing.
It's typically used to monitor bone marrow transplants to show as -- After the transplant, as that patient progresses, you can see whether or not that patient is continuing to make cells that are from the donor, which the -- which is the outcome that you want, or the patient is going back to making cells from themselves, which is not the outcome that you want.
And that's done, that monitoring is done over a period of time and it's done using RFLP testing.
So in this life or death situation of a bone marrow transplant, the same RFLP test that you've been talking about here today are used?
There was also some discussion, Doctor, about using these narrow windows to make a band match judgment. Do you recall that Mr. Blasier asked some questions about that?
Well, yes. He was asking about the window that you use to call a match and your calling it narrow. He was calling it wide. You use a window that's appropriate for your system.
Because the procedure that you're using does not allow you to figure out the exact lengths of the fragment in base pairs, that is, I can't tell you if something is exactly five thousand on a system that's used here. That's -- the system doesn't have that technical capability.
If I see a band and the computer imaging system estimates that it's 5,000, it could be 5,005 or 5,010 or 4,900. And, you know, 80, whatever. So each laboratory has to run multiple samples to assess how much variation you see in as a matter of routine.
And then you use that window, that plus or minus figure of how much variation you see routinely to assess how, whether two fragment size estimates are close enough to be considered a match.
Every lab that I know of that's doing an RFLP testing, we're talking about 50 or so labs, all use some kind of window to know whether or not they're going to call a match in the same manner that was described earlier.
Well, you do each band individually. You compare one band with the comparable band and the other sample and you look at those sizes and you decide whether they're close. You have -- occasionally they're identical, but are they close enough to be considered a match? And you do that and then you go on, do all the bands in the pattern.
And for example, in this case, you were looking at bands at five separate genetic locations; isn't that right?
And does using this system that you've described, affect in any way, the power of the RFLP system and the results that it generates?
No. This is part of the RFLP system. This is -- it's something that you have to understand in order to understand how to do the interpretations. And if you were to ignore this and ask that every band be identical in size, you would be ignoring an enormous amount of data and you wouldn't be doing an appropriate scientific job.
And once again, the same system is a system that's used medically and in diagnostic areas in a lot of other areas of science besides forensic?
Now, I want to touch briefly again on this question of item number 29 on the steering wheel.
First of all, Doctor, would you find it surprising if there was some DNA that shows up generally on a steering wheel in an automobile?
Cause someone else could have had a cut on their hand or been perspiring and left DNA at some other time?
Secondly, is it your testimony that based upon that little Department of Justice DQ Alpha type that he showed you, that although there are various possible combinations, as Blasier pointed out, one scientifically possible combination is a 1.3, 4, which would match Ronald Goldman?
There's a 4 that you can see. It's very, very faint. It's below the control dot. That means there may be another allele that you can't see and that other allele that you can't see can be any other allele that that system tests for; could be a 1.1, a 1.2 so and so on and so on. And it could be a 1.3.
Now, let's go to the polymarker test result that you got on Nicole Brown Simpson's reference file.
In your professional opinion; is that faint B lighting up the at GC cite a result of contamination?
It's a common occurrence. You can see it in our validation studies. You can see it in work from many other laboratories. You can -- I think it's even mentioned in the manual that comes with the test.
Finally, Mr. Blasier asked you some questions about the application of the product rule to the evidence samples in this case.
In your professional judgment was there anything at all improper about using the product rule in this case?
In -- and the sample sizes that the data basis are based upon, are those statistically proper sample sizes?
The amount of nanograms that were found in some of the evidence samples in the case, is it unusual to have a relatively low number of nanograms in a crime scene evidence sample?
And the PCR test is a test specifically designed to enable you to get results on low nanograms amounts; isn't that right?
Right. That's the whole value of that test is that you can use that test on samples where you cannot use any other test.
And the reason that the level of nanograms may be low at a crime scene as opposed to blood taken out of someone's arm is what, Dr. Cotton?
And the blood found at a crime scene is usually not something that is in a unique laboratory environment?
At least 99 percent is common to all of us.
If you were to ignore this and ask that every band be identical in size, you would be ignoring an enormous amount of data and you wouldn't be doing an appropriate scientific job.
I'm assuming that people don't intend to leave bits of themselves.
In my opinion, it's a result of cross-hybridization.