Dr. Cotton, with regard to these PCR results, do you feel it is appropriate to assign numbers to mixtures?
We felt that simply stating that these individuals were not excluded was also an appropriate way to report that data. When you are reporting data, there can be very legitimate differences from one laboratory to the next and there are some things where there is no exactly right or exactly wrong way to do it. That was the method we chose. There is also nothing wrong with giving a composite frequency for all contributors to that or all possible contributors to that set of types.
Just with respect--and I would like to direct your attention simply to the DQ-alpha results on item no. 78. Do you recall those?
All right. Do you have before you in one of your notebooks, the actual results, PCR-based DQ-alpha results?
All right. Your Honor, with the Court's permission I'm simply going to ask, and I think the DQ-alpha results are illustrative, ask the witness to write down were the summing process occurs, all the different types that would be involved in it, because I think that is probative.
Dr. Cotton, with regard to the DQ-alpha results, can you describe for the Court each of the various types that could have contributed to this sample, as the Court has expressed an interest in this type of summing process?
Well, we can give an example. If I miss a type somewhere in there, we will have to let me think about it, but I will get--I can give you the idea.
All right. With regard to the various types, would it then be the case--and first of all, what are the results on item 78, just DQ-alpha?
There is a 1.1, a 1.3, a 4 and there could be in there a 1.2 and you can't prove that it is there or prove that it is not there, so for purposes of this it would be prudent to include that as a possible fourth type, so let's include the 1.2.
Now, with regard to this summing process, are there then a number of different possibilities in terms of the types of people, of individual people who could have contributed to that sample?
Umm, no, I can't tell you how many right off the top of my head. And the other thing is that when you think about how many--wait. Why don't you ask me again, because maybe I'm not getting what you are asking.
Okay. With regard to the field of possible contributors, and I'm only referring to different genotypes--
--who could have contributed to that sample? First of all, are there more than two possible types that contributed to that sample?
Okay. When you say "many," do you have an approximation of how many without sitting down and actually--
Okay. Are there then a series of genotypes that may very well have absolutely nothing to do with the contribution of that sample?
You mean by summing all the possible genotypes that could have contributed to this grouping of four, is that misleading at all?
Well, let me ask it a different way. If one were to sum all of these genotypes to create a frequency, could you then end up with a frequency that is far more common than the truth as far as that mixture is concerned.
All right. By summing all of those frequencies could you then obtain a frequency that is far more common than the frequencies of the actual contributors to that stain?
Is that in your view--well, do you believe the summing process then has a weakness as far as being accurate?
The summing process gives you an extremely conservative idea of what percentage of the population could have contributed to these set of types.
KEY QUOTESo in this summing process are you including types that may have absolutely to do with item no. 78?
Yes, you are. There is something else that nobody has said here, and you--that is, when you do this summing process, you have to make an assumption, do I have two contributors? And then based on two, you would sum everything up. If you said, well, maybe I have three contributors and you summed from that perspective, then your frequency would be still yet more conservative, or if you postulated four contributors, so in doing that summation and doing that calculation, you have to make some assumption I'm going to do the calculation based on two possible contributors and then do it from there. If you did it based on three, it would give you a different result, and then in--also in doing it, you would need to say am I going to assume these people are Caucasian or Hispanic or African American and various combinations would also change the result.
In your view, scientifically, is it scientifically appropriate to simply conclude from a mixture that you are unable to exclude an individual or individuals?
I don't think it is misleading. Clearly if you say these two people can't be excluded, you might also be saying other groupings of people can't be excluded also. If you get my--that is, if I was asked can these two people be excluded, the answer is no, they cannot. Can other combinations of people also not be excluded? The answer would be that's correct, they can't be--other combinations of people can't be excluded either.
In other words, you feel to describe it in that manner, not just that the people can't--or certain people can't be excluded, but others can't be excluded as well, that that is scientifically appropriate?
The summing process gives you an extremely conservative idea of what percentage of the population could have contributed to these set of types.
There can be very legitimate differences from one laboratory to the next and there are some things where there is no exactly right or exactly wrong way to do it. That was the method we chose.
If I was asked can these two people be excluded, the answer is no, they cannot. Can other combinations of people also not be excluded? The answer would be that's correct, they can't be—other combinations of people can't be excluded either.
Yeah, probably more than ten.