Fourth Annual DNA Grantees' Workshop
Tuesday, June 24, 2003
MORNING SESSION
Question-and-Answer Session
MR. TAMBASCO: We've got time for questions. Anybody have anything before we break? Yes?
QUESTION: I have a question for Murray Brilliant. A number of us out in the audience are looking at things like mitochondrial DNA or ALU polymorphisms to infer biogeographic ancestry of genetic elements, and we see a lot of potential there. You mentioned earlier in your talk about a 1970s paper that was politically incorrect for one reason or another. I wonder if you could comment on how one distinguishes between a politically correct and incorrect investigation?
DR. BRILLIANT: Well, let me address the scientific incorrectness about the paper first. It assumed that African Americans are one genetic group and that the genetic diversity among them is quite high. Therefore, the assumption is that all the pigmentation loci would be identical throughout all African peoples, which may not be the case.
I think that the way this study was done in the 1970s, it probably had a lot of assumptions and I think some of the admixture was calculated. But I think the paper is useful in that it indicates three or four genes that contribute in a major way to differences in pigmentation. This study, then, is limited to African Americans. I think that when we look at other populations there are a lot of other ways to tweak pigmentation.
QUESTION: So are there particular ethical issues that we need to be aware of as we undergo this?
DR. BRILLIANT: I suppose so. My interest is not really in doing this to predict the ethnicity or race of a sample. I'm looking at it more in terms of the function of the genes themselves, and rather than come up with kind of a subjective way of classifying people, why not come up with an objective way. That is, that we measure skin reflectance rather than say this person belongs to this ethnic group or something like that.
QUESTION: Well, that said, is there anything wrong with predicting the ethnicity of an individual?
DR. BRILLIANT: It's just not something that my study is addressing. There are other studies that look for markers. They're called AI markers, ancestry informative markers. Clearly, that could be useful in many ways, but this particular study is just trying to determine more objectively how to describe a person in terms of hair color, eye color, and skin color.
There is quite a variety among the American population for sure and even within ethnic groups, so I'm trying to come up with more of an objective way.
MR. TAMBASCO: Jack, question or comment?
DR. BALLANTYNE: Dr. Brilliant, (inaudible). How are you going to look at these multi-genes (inaudible)?
DR. BRILLIANT: Yes, that's a very good question. Currently we have a smaller sample where we sequence axon regions and such, and then we look for polymorphisms that are, say, at least 10 percent in various reference populations. Those are the ones that we've been looking at initially.
But you're right, it's a large problem. For example, the P gene itself spans about 400,000 nucleotides and 24 axons, and some of the polymorphisms that we see that appear to be of value in this study are not coding polymorphisms. So perhaps they influence splicing and processing and/or time of expression and levels of expression.
Clearly, I talked about pigmentation in hair, eyes, and skin. There are timing issues involved in this, too, that lead to variation. For example, pigmentation in the eyes occurs early, and there is not really a turnover of pigment in the eyes. On the other hand, skin and hair are renewing, so there is more of a turnover in terms of pigment there.
Some of the controlling regions will certainly have a role. From these small reference sample, we're trying to locate some sequences that we think may be important, and they would be the more common ones.
MR. TAMBASCO: Roger.
ROGER: I'm wondering about color as opposed to reflectance and whether there are differences among people's skin tones if you look at more than one frequency. I'm just guessing how your reflectometer works. You said it was a single digit, a single number.
DR. BRILLIANT: Well, the number is an average. There are three different wavelengths that are emitted and the reflectance is the average of those three wavelengths. So it does take that into account. It's not just light and dark.
ROGER: But for someone looking for a suspect, he wouldn't be using a reflectometer.
DR. BRILLIANT: Well, we can translate it then back into the I through VI skin types. But actually, the reflectometer that we use is rather portable, so it could be used, I suspect.
DR. HERR: I want to ask a question about DNA degradation. We know the entire human genome now and the question is, is the degradation that you find random degradation or is it primarily due to restriction endonucleases at specific sites? In other words, is it primarily bacterially mediated?
The reason I ask this is, if we knew that it was primarily restriction endonuclease mediated, knowing the entire genome, we could predict those sites and identify sites to amplify, thus avoiding the restriction sites. Similarly, we could go back and look at all the samples of degraded DNA that you have, look at where the Ns are, and compare that to the genome to determine if it is primarily restriction endonuclease mediated.
DR. McCORD: I don't know whether I can provide one answer. There are a variety of different processes and there hasn't been enough study done. I think that's the real answer. There's been some work looking at effects, such as dimerization, methylation, and other things that are obviously not restriction related. But as far as the whole host of things that have been done, we weren't able to find anything in our brief search.
MR. TAMBASCO: George, comment?
GEORGE: If you look at some naturally degraded samples, you can see the (inaudible).
DR. FORAN: We did a study in my lab—it didn't work, but we did it anyhow—related to that issue. We let mouse tissues degrade, either by just taking out the liver and the spleen and various things and letting them degrade over time at various temperatures, as well as homogenizing the samples so that the mitochondrial DNA and the nuclear DNA would be in the same situation. Then we used real-time PCR to try to quantitate the relative amounts that the different nucleic acids degraded. We certainly found that mitochondrial DNA was degrading in its history in the whole tissue very differently than nuclear DNA and once you homogenized that, that changed.
We also found that different loci in nuclear DNA—for example, a ribosomal repeat and a single copy gene in mice—degrade very differently. What you said is very true as far as the ribosomal genes, which would likely not be histone-related because they're being expressed at a very high level and degrading much quicker than anything else. So there is different stuff going on within the nucleus.
MR. TAMBASCO: At this time I'd like to thank our panel once again.