DOCTOR CROW: Let me say the same thing that I said yesterday, and that is that we'll have an opportunity for general participation not reserved until the end, but after each of the parts of the program. So remind me if I don't remember this.

Since our regular speaker at the moment isn't here -- well, wait a minute. Here he is. It is my pleasure to introduce the late Doctor Ferrara, who will now speak to us.

PAUL FERRARA: Within your folders, all the members of the -- of the commission have a copy of a short synopsis really of our discussions at our last Laboratory Funding Working Group.
For those who don't have the benefit of it, with permission of the chair, I just would like to read the report into the record. The Laboratory Funding Work Group. Now that consists of myself, Woody Clarke, Dave Coffman, Cecelia Crouse, Barry Fisher, Steve Niezgoda and Dick Weedn. We continued to develop or continued to develop the costs associated with the elimination or reduction of crime scene evidence backlogs nationally. The estimation of the volume of such samples from past, present and future cases is in development through ongoing surveys of laboratories and law enforcement agencies, such as BJS, NIST, PERF, CODIS, and ASCLD, the American Society of Crime Lab Directors.

This data should be available for review and analysis by the Laboratory Funding Working Group by the fall of this year.

Meanwhile, the Laboratory Funding Work Group continues to refine the costs associated with the collection and preservation of convicted offender data bank samples. The work group believes that the consideration of preservation costs is important to assure that all convicted offender samples are maintained in a secure and protected environment so as to assuage any privacy concerns.

The Working Group recommends the preservation of these samples on three levels: scientific, legal and practical.

Under scientific grounds:

(1) The Laboratory Funding Work Group, like the Research & Development Working Group, believes that some of the most exciting current developments would not have been predicted ten years ago. In order to maintain flexibility to take full advantage of the future technological developments, retention of original samples is necessary so that state data banks can retroactively apply advanced technologies to stored convicted offender samples. In the past ten years, we have already experienced one transition in methodology, which necessitated reanalysis of convicted offender samples.

(2) As new loci are identified, retained convicted offender samples, without personal identifiers, provide a source of random samples for establishing allele frequency distribution data for the DNA advisory board mandated population studies. In addition to population frequency data, analysis of these samples also provides important data regarding potential off-ladder and microvariant alleles.

(3) Good laboratory practices dictate that the retention of convicted offender samples provides an additional quality assurance measure for the detection of any potential sample mix-ups.

(4) Retaining -- retained convicted offender samples provide a well-defined source of previously analyzed known samples, again without personal identifiers, for use as training, validation and internal proficiency testing samples.

From a legal standpoint, with all due deference to the legal community, when the DNA profile of a sample from a crime scene is found to match a convicted offender profile in the data bank, that match becomes the probable cause for an arrest or search warrant to obtain a new blood sample for confirmation. At trial, the chain of custody for all samples, as well as the samples -- as well as the sample themselves becomes subject to subpoena. In fact, a provision of Section 2.b.2.c of Senate Bill 903, the Violent Offender DNA Identification Act of 1999, states that "for criminal defense purposes, to a defendant, who shall have access to samples and analyses performed in connection with the case in which such defendant is charged..." Any of a number of definitions of evidence in Black's Law Dictionary could be applied to convicted offender samples.

States are, or should be, treating standard blood samplings as evidence, complete with appropriate security, chain of custody and supporting documentation. Data bank analysts have already been subpoenaed to testify to their analyses of data bank samples in the same manner as those who performed the crime scene work. The costs and resources necessary to provide this assurity to the courts are not insignificant or unwarranted. In point of fact, it may be required by state law; for example, in Virginia, to quote Virginia law: "The remainder of a blood, saliva or tissue sample submitted for analysis and inclusion in the data bank...may be divided, labeled as provided for the original sample, and securely stored by the Division in accordance with specific procedures adopted by regulation of the division," in this case they are referring to the division of forensic science, "to ensure the integrity and confidentiality of the samples. All or part of the remainder of that sample may be used only (i) to create a statistical database, provided no identifying information on the individual whose sample is being analyzed is included; or (ii) for retesting by the Division to validate or update the original analysis."

This is an excerpt from the Code of Virginia. I might -- I might add that that law was passed in 1989, ten years ago.

The Work Group, therefore, considers destruction of these samples after analysis, which analysis may occur years after collection, to be inconsistent with the laws governing preservation of evidence.

Finally, from a practical standpoint, costs associated with the collection of convicted offender samples have been discussed at length by this Working Group. Emphasis on funding and personnel time are critical components. The documented and complete destruction of these samples, which would constitute of course being the biohazardous waste, would only further add to the cost of the laboratories; resampling convicted offenders when a technological improvement is necessitated is an unacceptable alternative.

Having thus established a set of reasons why convicted offender samples should not be destroyed after analysis, the Working Group then turned its attention to any possible advantages of sample destruction. None could be determined. The Laboratory Funding Working Group perceives these samples when treated as evidence to be eminently more secure and less prone to misuse than the myriad of biological samples with personal identification that the general public willingly provides everyday to blood banks, doctors' offices, clinics, employers and insurance companies without any thought of how these samples are used and what, if any, regulations govern their use.

The Work Group instead recommends that states adopt regulations that require convicted offender samples to be maintained in such a manner that is consistent with privacy and confidentiality concerns.

And that concludes our report, and I will be glad to...

DOCTOR CROW: Yeah, well, thank you, and it's open for discussion.

DOCTOR REILLY: You have almost, but not quite, convinced me. I would like to return to -- let's begin. I would like to return to the scientific criteria and just ask a few questions.

Let me begin with item two, saying it provides -- retention of the samples provides a source of random samples for establishing allele frequency.

Certainly, we could establish allele frequency from another collection of samples.

PAUL FERRARA: That is correct.

DOCTOR REILLY: It would be -- I don't know -- what size -- what sample size would you need to do that? Would you want a hundred, a thousand?

PAUL FERRARA: I put 500, a thousand samples.

DOCTOR REILLY: So I must say that, from a matter of principle, I regard that as a trivial argument in favor. There are just databases all over the United States from blood banks. I could get 500 samples for you in 24 hours at almost no cost. So I regard that as not a persuasive argument.

Okay. It sounds like you are not in great disagreement with me?

PAUL FERRARA: That is right.

DOCTOR REILLY: There is other sources?

PAUL FERRARA: That is right.

DOCTOR REILLY: Okay.

PAUL FERRARA: No disagreement there.

DOCTOR REILLY: Item three, good laboratory practices dictate keeping the samples to check on mix-ups.

Now I think that is a little bit more persuasive from the point of view that we want a system that has zero tolerance for errors, essentially. Nevertheless, in all of these matters, I think you have to weigh, and we will get into it later, and I have some discussion points upon it later this morning, you have to weigh the benefits of doing that against perceived, and I want to underline perceived threats to individual privacy. Assuming that you are operating in a system that is somewhat greater than 99 percent, hopefully in its accuracy, and you have ways of checking that through the use of random spiked samples, where you can do validity checks, I'm not sure how powerful that argument is. I agree that there is a potential for, if you will, miscarriage of justice here that would ultimately be captured when you reanalyze the case.

Similarly, four, retained convicted offender samples provide a well-defined source of analyzed known samples. It seems to me for training validation and internal proficiency purposes, you could easily use other kinds of donated samples to do that.

Would you agree?

PAUL FERRARA: That is correct.

DOCTOR REILLY: Okay. So let's go back to number one. This is actually the one that troubles me the most, and I'm saying this not so much because I disagree strongly with your group, but I'm imagining the arguments that will be made from other corners of our society in response to this document.

We live in a world in which genomic technology is moving very, very rapidly, you and I both know that, and there is much discussion about where it will go with snip maps and, you know, also sorts of different things. But the fact remains, as we heard in painful detail yesterday, that we are light years away from having enough money to work with the technology that we have now. And it seems to combine that fact with the fact that I actually believe that the STR system that is going into place now is a very powerful system, and it is unlikely in my mind that a system will emerge that will be so much better in the next five to ten years that it will become an overwhelming technological argument to employ it. I really don't buy, and I think a lot of people will not buy, the argument that we should retain the system, retain the samples, because in three years we will want to start all over. That given the funding problem seems not to wash at all in my mind.

PAUL FERRARA: Let me address -- let me address a couple of the points you have raised, Phil. We have -- we have now been operating a data bank for ten years in Virginia. We are talking about a database right now approximately 200,000 samples, 75, 80,000 of which are actually profiled and in the state level of CODIS. During the course of collection of samples from -- at the prisons, at the jails, the transfer of the samples to the laboratory, excisioning in the laboratory and then through the various steps of the analysis and then the incorporation or uploading of the resultant data to CODIS, I have to admit that mix-ups have and do occur.

Now, I would also agree with you, Phil, that that percent is extremely low considering the number of samples; however, it has given us enough concern that when we make a match using the database to a convicted offender by comparing a crime scene sample, the first thing we do when we make that hit is determine if the individual we have identified is free or currently still incarcerated on that charge, or some other charge. If we determine that the individual is still in custody, we take the extra day or two before informing the law enforcement agency that we may have a hit and that we recommend that agency go out and investigate this individual, we go back to the original data bank sample and rerun it.

Now, we've ruminated about this in Virginia for some time. It's one of these steps that considering from a cost/benefit standpoint, I'm almost inclined to say it's not worth doing it. The likelihood of a sample mix-up of this type is extremely remote.

The other side of the coin is we were concerned and remain concerned about giving a law enforcement agency the name of someone who we have identified and then to find out that because of a sample mix-up of some sort that the person we have identified and sent the police after have gone, used as probable cause to arrest somebody, take a blood sample, God only knows what events might occur during that particular arrest and search warrant so as to avoid that we thought we would take that extra step.

We have also encountered countless numbers of different situations that have asked us -- required us to go back to the original database samples. We found out, for example, that in Virginia, we have at least four sets of identical twins in our correctional system. We would not have been able to determine that had we not been able to go back to the samples and retest them.

With respect to the chain -- to the changing technologies, I mentioned at least one change. That was specifically from RFLP to PCR, but in between there, we ran through a series of changes in either -- in either methods of detection of the alleles or new genetic loci. Right now, in the U.S., you have a series of various types of technologies being used in the data banks: RFLP, PCR-based loci. Often what can happen is you are looking to do a search of someone's database or search a particular sample, and you find that because of the different technologies in use, it's useful for one or both of the laboratories to go back to the original sample to check at additional loci.

Now, with respect to the 13 loci when you come to situations that we often encounter in the laboratory of mixed stains, you will find that often even 13 loci are not enough to develop the degree of specificity necessary to conduct a database -- conduct successful prosecution, because what happens is you are only able to use a finite smaller than 13 number of genetic locations.

In those situations, we have gone back and looked at other loci available to us, such as FFFL, for example, when other -- when other use all loci are possibly available to us.

While I admit that there is ways to cope with the destruction of samples, I think the Working Group really felt that given all of the various scenarios, both already identified and that we can't even consider in the future that where it would be advantageous to have the original sample, after all is said and done, are not these genetic data banks, forensic data banks the least concerned for the general public for purposes of privacy and confidentiality compared to the rather haphazard form that samples are -- biological samples are provided.

DOCTOR REILLY: Thanks.

If I could respond to you. I have no doubt that from a technical point of view there will be instances such as those you have identified and others in which it will be helpful to a particular case to have access to a storage sample. I grant you that. For me it's a balancing test against how the society may perceive the commission's, if we so recommend, decision to store whole DNA. And I'm not sure that they will weigh it the same way your Working Group has weighed it. As a matter of fact, I would say it is actually dangerous to submit a report that says you could find no reason at all in favor of destroying samples. I can guarantee you, I could have assembled any number of committees that would have come out the other way, are very reasonable people.

The fact that -- the fact remains that in human subjects research in the United States, in clinical medicine, throughout the United States, as technology becomes more powerful, and you see it everyday -- it's on the front page of the papers today -- there are concerns about privacy. I absolutely agree with you that the odds of this database being violated are much smaller than the odds of my doctor's office leaking information about me, but I don't think that is the issue. Theissue is that these databases, these forensic databases, are emerging at a time when the public perceives correctly genetics to be extremely powerful information, and if there is a least invasive alternative, they are going to prefer that, the least invasive here or the more protective alternative.

For me the case hasn't been made yet, and I wondered tactically, even if you and your committee are absolutely right, if it still is not better for the future use of this technology in society to say, hey, you know, even though there are benefits to retaining the DNA, we see the concerns about dignitary harm to privacy, and we are willing to go forward without it, because it seems to be the actual cost to using the databases, the criminal justice system would be very small in not retaining the samples.

PAUL FERRARA: Could perhaps we look at it from this standpoint, Phil, that here is an opportunity to demonstrate to the general public, all of whom have concerns about privacy and confidentiality, that we sort of -- forensic science data banks can serve as an example and a leader in terms of how data banks of this nature should be maintained to provide that degree of assurity, in other words, set an example. Clearly, this question about confidentiality and privacy of genetic information, and I'm out of my league here, but it has to be -- will arise with regards to so many collections of samples; and if this -- if these data banks here are the first to really address this issue, can that serve -- can we serve as an example of how this data can and should be protected.

DOCTOR REILLY: I think there are others who want to speak, so I promise to shut up after this, but let me ask in closing: Do you feel the currently existing statutes offer the appropriate level of reassurance to the American public of the degree of security to the database that you would like to see conveyed?

PAUL FERRARA: Not in all states, no; in Virginia, yes, but not in all states.

DOCTOR CROW: Michael first. Then I will call you, okay.

MICHAEL SMITH: Well, I have, I think, two questions. The first is I was following the discussion about the disadvantages of not having biological samples available later as disadvantages of the prosecution of a case, but we don't prosecute samples, so I am still puzzled why say identical twins? Why do you need for the purposes of prosecution the identification sample on which the probable cause for arrest was based? So that is one.

The second one is --

BARRY SCHECK: Michael, your question is to Paul --

MICHAEL SMITH: Yes.

BARRY SCHECK: -- and I think it will occur to all of us sitting here is that every one of the practical considerations that you raise, name one where you couldn't solve it by just going back to the person incarcerated and getting resampling. I didn't hear one.

MICHAEL SMITH: Yeah.

PAUL FERRARA: In two of --

DOCTOR CROW: Maybe Mike had a second point, if you want to.

MICHAEL SMITH: I think Barry makes a good point. I will have my time for the second point. We need to answer this one.

PAUL FERRARA: Consider the scenario where you're entering a large number of genetic profiles up into CODIS, and as you do so, you search those profiles for any existing match, either with other convicted offender samples or with crime scene material; and when you get a -- when you get two individuals with identical matching patterns, and different names, and different Social Security numbers, we as a laboratory are sitting there saying, oops, what happened here? How do we ascertain that, in fact, these were identical twins? Well, the first thing we did is go back and run those samples to make sure that indeed the profiles were identical. It was informational for us to determine not only in this particular case that these individuals were identical twins, but in other cases where there have been sample mix-ups, it has allowed us to determine at what stage that mix-up occurred, because I will admit another thing. I will concede another thing, and that is that even though we go back and resample the original sample, that doesn't take care of us if a mix-up occurred during sampling, but again we at least have narrowed down where that mix-up may have occurred. It seems situations where a sample, you know, was sampled twice and put in two consecutive wells, and then we have to ascertain and make sure that the whole string of following samples and wells is not misplaced by one. These investigations have all depended greatly upon our ability to go back and have those original samples.

BARRY SCHECK: But in that instance, and in all the other instances, you have to go back to the original people to resample them to see if they are identical twins.

PAUL FERRARA: No. Go back to the original sample.

BARRY SCHECK: Oh, you mean if there was a mix-up that occurred?

I mean you are saying, well, I went back to the original samples to see if we made a mistake that matched them, but your investigation didn't stop there. You went back to the individuals and resampled them to make sure they were identical twins?

PAUL FERRARA: No, what we did then is go to the correctional people and get the information that a -- that confirmed that they indeed, you know, investigated.

BARRY SCHECK: But in answer to the question, you could have -- if you had thrown away the blood samples, and you had typed them and you found you had two identical samples, you could have gone back, found those individuals and resampled them, right?

PAUL FERRARA: Yes. Yes.

BARRY SCHECK: So far in all the examples, and I mean this is --

PAUL FERRARA: Oh, yeah. No. No.

BARRY SCHECK: And I know you are fair about this. Every example you gave us, unless you hypothesize the sudden death of somebody, right, which would be a real rare event, you could solve each one of these problems by just going back and resampling the individuals, yes?

KATHRYN TURMAN: It happens in D.C. on a weekly basis.

DOCTOR CROW: Let me call on you first.

DAVID COFFMAN: Well, first of all, I would like to -- I just want to respond. I have something to read into the record later, but I do want to respond that sudden death -- I am not even going to say sudden death -- death of inmates is not a rare event. We have inmates that are dying all the time from AIDS and other diseases in the prison system. So that is not an absolute rare example.

DOCTOR REILLY: And it doesn't terminate the right -- the ability to get DNA though either.

DAVID COFFMAN: If they died six or seven years ago, and you determined a problem --

DOCTOR REILLY: If you really wanted to know, you could exhume them.

DAVID COFFMAN: Well, that is a little --

DOCTOR REILLY: It has been done many times.

PAUL FERRARA: We do that constantly.

DAVID COFFMAN: Oh, I know, but as far as an example, someone asked for an example where you could necessarily easily. I know all these things you could go back and do it, but is it feasible? Is it time productive?

For instance, we have a low stringency search request. Not a low stringency. There was a case that was done in 1990, very minimal sample. They were only able to get four of the genetic markers. Only two of those genetic markers were the CODIS core loci, which the database was made up of. You could get several hundred, even several thousand matches back. You are not going to have the investigators go out and draw all these individuals, who may have served their time -- they are not under custody anymore -- and have them all retested to confirm that match. If we have it on file, we can do it and only inconvenience the person that is the most likely person who committed that crime.

And as also in our -- is in other states moved to all felons, or even I am hearing some states going to all arrestees. You are going to have people who are not in there for necessarily violent crimes, and there is a stigma attached to someone who is connected to a sexual assault; and to sit there and start rounding up thousands of people, because you had a search that didn't have the, you know, the core loci, I think that is very cost prohibitive, and it's -- I think it's better to maybe just regulate safeguards into the system to prevent this kind of thing.

DOCTOR REILLY: Well, I have heard a lot of things being cost prohibitive in the last two days, and that is about 19th on my list of -- I mean we can't even process the samples we have now.

I am asking you, though, since you were looking at me when you were talking, I am asking you to remember what I said. I am asking you to weigh the arguments that are made here in favor of retention in a public that is not predisposed necessarily to accept a long-term storage of whole DNA on potentially millions of Americans; and if anything, if you think deeply about it, you could argue that I'm speaking out to defend the system we are trying to put in place so it doesn't get harmed by an overreaction to the public.

The first time that there is a report that somebody leaked whole DNA, or sold it, or did something crazy with it, there is going to be a big backlash against this.

DAVID COFFMAN: I understand that --

DOCTOR REILLY: Okay.

DAVID COFFMAN: -- and I do believe we need to have the safeguards in place, but let's also make it clear that we are not talking millions of Americans, such as grandma and grandpa or Aunt Nellie. We are talking people convicted of felony crimes.

BARRY SCHECK: Maybe.

DOCTOR REILLY: We started with sex offenders, but the net is being cast very wide, very quickly. Once a criminal, not always a criminal, I hope, in our society.

DAVID COFFMAN: Well, 67 percent do reoffend. That is a known statistic.

DOCTOR REILLY: Therefore, 33 percent don't.

DAVID COFFMAN: Right. Well, that is the choice they made when they committed the crime, in my opinion, also. And also I am a little concerned that people are only concerned about five or ten years out. That is not very long at all. That is not -- that doesn't make me feel good about destroying the samples, because you are condemning our nation's databases to start over ten years, and we have already been in existence like Virginia, we have been in existence since 1989, and you don't start collecting samples and just start making hits and helping solve crimes. It took seven years into the process before we started getting any real benefit from our databases, andit's growing each year, and it's because we built our databases.

DOCTOR REILLY: Pulling out whole DNA is not the same thing as expunging the computerized code of the 13 STRs.

DOCTOR CROW: Jim.

MICHAEL SMITH: I have got a second question. Well, it's a related point so it does seem to me not inappropriate to dump it in this question.

I noticed that the conversation has more to do with the database, its construction, maintenance and reliability to those who are managing it than with the prosecution of a case.

Now, the second point is it really feels, I guess, that is a lot of the concern that people might have is not about the use of DNA for identification purposes. The reliability of databases expressed in their exclusive focus on identifying alleles, or whatever the code is, system seems to me to be important. Now if the record retention is of that identification profile, the -- it's easy, I think, to argue against efforts to constrict the use of the technology. If, however, people have reasonable fears that other uses will be made of the DNA that is retained, and I think they do, then it seems to me that it's much harder to resist attempts to constrict the use of the technology.

Now, this is already evident in the questions about whether or not arrest samples should be taken. And to the extent that the future is what we are concerned about, it seems to me to lock ourselves in to convicted offenders is not necessarily wise, and that in a sense you do that by retaining the samples, because it's very hard to explain to somebody that you can reliably predict that nobody will make future use, forensic or otherwise, of samples that are on file with the government. The government is not that widely trusted.

DOCTOR REILLY: Indeed, the statutes permit it on anonymous research.

MICHAEL SMITH: The state statutes, yeah.

DOCTOR ADAMS: The point has been made, and I won't belabor it, but Barry you made a comment that you didn't see anything within the practical portion of Paul's report to really warrant the retention of samples that have been collected, but I think in the first paragraph when it talks about resampling convicted offenders because of changes in technologies, in the first paragraph under scientific, which talked about the same thing, I think it's shortsighted to think that STRs will be in existence forever. In fact, within the last ten years, the technology has changed more than once. It has gone from RFLP to various forms of PCR based to now STRs. So to say that STRs will be here forever, that is shortsighted; and ten years --

BARRY SCHECK: We agree.

DOCTOR ADAMS: -- and ten years is not very long at all for other changes to be made.

JEFFREY THOMA: Just like RFLP is still available now, even though you don't really want to use it in that much of the future, if you are going backwards, you can still do the STR match to see if it's within the database, and then use whatever technology has developed since then. So STR is used as the database system.

DOCTOR ADAMS: That is really not as easy as it sounds. Laboratories don't turn on and turn off technologies at a whim. Laboratories have to be proficiency tested, have to undergo accreditation in these areas and can't simply decide next year to retest using RFLP when they haven't been proficiency tested in that for the last five years.

PARTICIPANT: Oh, could I address this actually?

DOCTOR CROW: Oh, yes. Let me get Woody first, and then I will call you.

GEORGE CLARKE: Just a couple of observations. I want to return briefly to the -- what I think was identified, or at least described as a trivial aspect. The allele frequencies -- and actually I don't think it is quite that trivial, and I want to jump into the courtroom for a minute, since that is where some of these fights are. And unless you have been in the -- that has been in the midst of an attack, and I mean an attack on the allele frequency data, the probabilities, estimates and so on, and while scientifically, perhaps, it is trivial or is trivial, you know, is 500 samples enough, and scientifically that appears by all accounts to be true, that is different from whether or not that is trivial to 12 individuals in a courtroom. And when you hear, for instance, a population estimate that these samples match in only 1 out of 500 people have those characteristics, it's a very common attack where you have only sampled 500 people, haven't you, and jurors don't like that.

So, again, while scientifically it may be very trivial, as a prosecutor in a case where DNA is being attacked, you bet your life I want more than 500 people in that database. I want 10, 20, 50 or one hundred thousand people in that database.

DOCTOR REILLY: It looks like you are speaking to me since I characterize it.

GEORGE CLARKE: I am not speaking right at you, but I did look at you Phil, that is true.

DOCTOR REILLY: You misunderstood my point, and I have another solution to your problem. First of all, what I meant was it was trivial to get another set of 500 individuals outside of the database. That is trivial.

GEORGE CLARKE: I agree.

DOCTOR REILLY: Blood banks provide information.

Secondly, there are a myriad of activities going on now throughout the world that would lay to rest forever those issues; for example, the Human Genome Diversity Project, which is amassing allele frequencies on many loci across virtually all populations around the globe in far larger numbers than 500, for example.

GEORGE CLARKE: It's just not the way it goes in the courtroom.

DOCTOR REILLY: Well, I haven't been in the courtroom in the same way you have, but I have been there a lot myself, and I know that, but it seems to me that again, and I am asking you to weigh things that again is not an argument that supports in my mind convincing the American people to store whole DNA.

GEORGE CLARKE: I agree. It's certainly not the most important aspect, but I don't think it should be dismissed.

DOCTOR REILLY: I don't dismiss any of these things that were raised, but I don't think they add up to an overwhelming argument.

JEFFREY THOMA: And whether you still have the data base, even though you don't have the whole samples. I mean you still have that.

GEORGE CLARKE: I understand that, but in any event, that was actually the more minor point.

DOCTOR CROW: I understand you have the 13, but I would like to hear some discussion of how badly you would need to add additional loci to the 13 or snips or mitochondrial DNA.

Do you want to answer it?

BARRY SCHECK: How badly one needs that is how badly one needs to do --

DAVID COFFMAN: To move beyond the 13 STR system.

BARRY SCHECK: Let's step back from this. I think I'll answer that point. We have all -- we all understand these problems pretty well, and it seems to me that what happens here with your group, and I think you raised from an institutional point of view almost your charge, tell me all the different possible reasons one might want to retain the samples, okay, without -- and I don't think it was really your charge to sit here and weigh the privacy interests and even the political perception of data banking, because you have freely admitted, I think, that, you know, so many of these things can be done with resampling with comparably trivial costs and that the real argument here is the one that Dwight identifies, number one, what is the real cause, you know, we may within ten years come up with a new system, and we are going to want to resample everybody at that time, right?

Or your point, Doctor Crow, what about mitochondrial or snips, to put it concretely; and, yeah, you know, if we do within the next ten years decide that we need these new markers, and we have to go out and, you know, resample everybody to do it, that would be an additional cost that could be saved, right, if we hadn't destroyed the samples. I'll give you that. And that is to me the only real argument, but I'll tell you why it fails and fails badly in my judgment, and you are missing -- you have got to step beyond the really small institutional concerns, because you really haven't given us, other than this potential future point, any serious bureaucratic reason not to destroy the samples.

As Michael was saying before, you go out and you ask people when you take their DNA, all right, trust the government. The government says, we'll only use it for identification purposes, and that is only in the states that have such statutes. What about all the other states that don't have those statutes. What about this statute where they say they want to use it for, quote, charitable purposes. That can mean anything soon. And the political instinct of most people is I really don't trust the government with my DNA. If it is there, somebody will find a way to get it for other purposes than identification with all the technology available. That is a visceral gut point with the public.

And, Paul, I agree with you that the ultimate objective is to persuade people that we can operate these databases in a secure way without bringing on brave new world. That is really our objective. That objective is better served by destroying the blood and taking the air out of the balloon that there is any possibility that these databases will be misused, and that is going to be become increasingly important.

Look at it, we have just spent a day talking about the enormous funding gap between what we can do with the technology and what resources are available in California. That was terrifying yesterday, wasn't it, they have three analysts for Los Angeles. We all looked at each other like, you know, what world are we living in? You want funding from the public on this then you really have to demonstrate that you can be trusted, and you can be trusted, and your best argument is we threw away the blood, because it's always going to come down to that. That is peoples' gut fear, and we are really eventually -- and eventually we are, in the next two or three years, you are going to be asking for billions to make this work.

Not only that, to make this work effectively, you are going to have to go out to deal with the really tough issues that we don't discuss very much, and that is what about these DNA dragnets, what about asking people for elimination samples. People are not going to give you elimination samples if they know that you are also storing blood and that there is all these other issues raised. You have to really build trust, okay. And so it seems to me actually a critical point for us, if this commission comes out and says, you know, there are some potential down-the-road costs here, and there may be some possible instance that we can hypothesize, you know, a dead inmate that we can't exhume, and we can't find his parents; therefore, we can't solve a, you know, a technical problem by resampling. And really that is what you are talking about in the final analysis.

The political game that you are going to have by coming out and saying, you know, it would be preferable to destroy the blood samples so that there will be no argument, you are going to get so much many more benefits. It's really the right thing to do.

PAUL FERRARA: But to do so isn't that contributing to the misinformation that the public would have regarding these data banks?

I mean isn't it suggesting that no, you can't trust government?

Isn't it saying that these forensic scientists are really geneticists in disguise, who could possibly even do anything of a genetic nature? I mean, you know, it's all we can do to knock out 13 genetic loci. I mean we are not -- again it's --

DOCTOR REILLY: Can I respond to that. I think that is an important point that you raise in response, Barry.

Let me ask you this question for the sake of argument. It relates to something in that little memo I wrote. Would you, as a trade off -- not that I have a right to trade off anything -- as a trade off in exchange for retaining whole DNA would you be willing to amend all state statutes to eliminate research of any kind on the samples and a criminal provision for any misuse of sample, a significant criminal provision that will lead to conviction and having your own sample put in the bank?

PAUL FERRARA: Absolutely. In fact, that would be my recommendation. I mean we feel that the statute is like Barry's.

DOCTOR REILLY: But I must say that although I appreciate your point about contributing to the argument, I agree 100 percent with Barry that in the larger context of the society, unfortunately these databases are unfolding at a time when there is great concern about employment discrimination, insurance discrimination. I actually don't think there is much of that either, and I have written a lot about that, but the perception is more important than the reality.

DOCTOR ADAMS: I mean it would be different, wouldn't it, if we could be very specific about what the future need is. It's because we are not specific, that is, we don't know for sure exactly what that future need is. It opens up the imagination to other kinds of needs that you don't actually have and wouldn't ever, but that is part of the difficulty.

DOCTOR REILLY: I must say though I find myself -- I find it uncomfortable to be justifying retention of samples in anticipation of a new technology when we are so far from being able to apply the technology in place.

PARTICIPANT: Right.

DOCTOR REILLY: I find that such a hollow argument.

BARRY SCHECK: I mean how many of you -- you know what is really interesting? Many of us sat around the table at the first NRC, right, and Dwight, when the FBI was justifying DNA and DNA data banks, what was their first argument? I will go back, and I will show you all the crime lab digests. You said, generic with the FBI, well, we only took junk DNA, and all we have in Washington, D.C. are these autocrats. That is all we have. We don't have the blood. And that was the argument. All right.

Now, of course the states have the blood, and we all recognize that is the issue, but I am only pointing out that any time you debate this with anybody, and you try to justify the databases and say it really serves a good purpose, and I am the person that goes out and talks with all the people, you know, from the Civil Liberties Union and everything else and saying no, no, no, no, no, and I say look at our statute. We wrote our statute so that it has criminal protections, and we protect it and everything else, and they say, I don't trust you. I don't trust the state with the DNA. How do you know we won't have another eugenic scare? How do you know we won't have another one of these lunatic things like sickle cell anemia? How do you know that they won't go after violence genes, sex offender genes for research if it's there. Tell me that in ten years, like you say, the technology will change, there won't be a sudden desire to grab that for another purpose. And, you know, you really can't win that argument. You can't win that argument in the long run. The only way you win that argument and get more money for this technology to be applied correctly is to destroy it and say, listen, it's impossible, because we have destroyed the blood, and there really isn't any justification here bureaucratically that couldn't be taken care of with resampling, or couldn't be, as Phil said, and I think we have to agree, his use of the word trivial in this context is correct insofar as you can go to other sources to get allele statistics for frequencies. I mean it's not hard to do that. You don't necessarily need this. I'm telling you, you are making a terrible error if you don't cease this opportunity, this commission now, and make this a recommendation for the states. You'll be so far ahead of the argument.

DOCTOR CROW: Woody.

GEORGE CLARKE: Well, just as an observation, I suppose, as much as anything, and I want to see if I understand your suggestion, Barry, by destroying the samples and assuming that a new technology develops that then becomes used routinely on criminal casework, would that then require going out finding those people and collecting again?

BARRY SCHECK: No.

GEORGE CLARKE: All right. Then those people are left out of any database searches; is that right? Is that the suggestion?

BARRY SCHECK: We have got them on STRs.

GEORGE CLARKE: Oh, no. No, what I mean is when they have been sampled and let's say profiled using STRs, and that when a technology, and I don't know that there is frankly much doubt that there will be new technology --

BARRY SCHECK: Of course.

GEORGE CLARKE: I mean whether it's chip based, or any of the other suggestions that we have already heard, I think we all realize that in one form or another that is a fait accomplit. But my question is those individuals, whose samples who have otherwise qualified have been profiled using STRs and are then destroyed, do those individuals then need to be picked up in a new dragnet? That may be the wrong term, but a new collection?

DOCTOR REILLY: I think it is the wrong term; but nevertheless, let me respond to that, and the first response is actually a dangerous one for you to make, because I know that there will be many who will immediately criticize it. But if the technologies are going to advance as much as you anticipate, and I think they will, the issue of transferring them is very different. There is no reason why they couldn't be run in parallel.

DOCTOR CROW: I am having a hard time staying out of the argument so I will. But I will ask a question. I think the distinction is being made between existing technology and future technologies is the wrong distinction. Many of those future technologies are here right now. They are pending nuclear-type probes that are being sold. There are -- I have already said snips and mitrochondrial DNA. These are not distant future technologies. They are here now, and the question is whether they are useful enough to make it important to retain samples beyond the 13 loci.

DOCTOR REILLY: That is certainly true, but, you know, I am wondering, and I am not necessarily the person to answer it, but I'm wondering how much more persuasive Y chromosome sequencing, for example, and David Page has sequenced the Y chromosome, most convicted felons and men you could just have a little Y chromosome sequencing chip. I am wondering how much better those technologies are than STRs. Why not say, which is, I think, a perfectly valid argument, STRs are good enough. We will make a collective decision in the interest of cost to keep in place the system we have, 13 STRs for X number of years. It doesn't mean that it's there, we have to use it.

DOCTOR ADAMS: We said that about RFLP though ten years ago. That was the system of choice. That was the best technology available at the time, and yet it changed.

DOCTOR REILLY: There are a couple of differences, I think, Dwight. One is that at the time the number of existing samples, the database was trivially small.

DOCTOR ADAMS: Right.

DOCTOR REILLY: Okay. Secondly, I think the advance from STRs or RFLPs is far greater than maybe snips over STRs in terms of the yield. I don't think that it is a clear distinction that one is that much better than the other.

I mean people love STR technology, as far as I can tell. I am not hearing somebody say, don't use it. Let's wait for chip technology. Who is saying that?

DOCTOR ADAMS: One way to understand this is to say that the technological challenge for the identification database purposes would be should a new technology for identification purposes emerge that we prefer the technological challenge would be how can we continue to access the old 13 STR database. The answer to that question would be important to get, but I think that Phil and I and Barry anyway are assuming that that is the technological challenge that needs to be overcome and would be overcome, and it would be worth taking that challenge on to get over the political and legal hurdles of retaining whole DNA.

JEFFREY THOMA: And even without that argument, Dwight, your argument that RFLP is gone, and we thought that it would be here for a while, you are always going to need RFLP for appellate and habeas purposes anyway, even though you are saying, you know, they have to be proficiency tested to even use it, but it still has to be available. The system has to -- in the legal system, it has to continue. I mean STRs would always have to still be available if they were used for any purpose, even if just databasing. And I agree with what Michael is saying and what Phil are saying. We can say we can stay with STRs for at least this purpose of the database; and if we go further, as Jim says with snips, or whatever we go to next, we can still revert to STRs. One of the reasons we have STRs now is the low cost and the efficiency of being able to do it and compare it. That is it is going to take a little while even if the technology advances to overcome and get down to where it's more efficient cost wise and otherwise to use another system.

DOCTOR CROW: Let me ask Dawn Herkenham if she has something she would like to say.

DAWN HERKENHAM: Yes.

DOCTOR CROW: I thought so.

DAWN HERKENHAM: Hi. I am Dawn Herkenham with the New York State Police, and I know that Doctor Reilly had asked a question about state DNA database laws, so fortuitously, I think the FBI had passed out something on DNA database laws, and I have listened to a lot of discussion, and I think there may be a misunderstanding that only a handful of states do have restrictions on disclosure, and there are actually 46 states that have restrictions on disclosure of DNA records, no disclosure of DNA samples. There are 32 states that penalize the unauthorized disclosure of DNA records; and for the most part, DNA records are defined within the state laws as the results of the DNA analysis or the DNA profile itself.

So there really are not a lot of statutes out there that would wholesale allow a crime laboratory, or whoever has possession of that sample, to disclose it to another entity. And, obviously, you don't have to take my word for it. What we try to include in the FBI's publication under the provision access and disclosure was actually verbatim the language in that particular statute. So you can sort of read that for yourself.

MICHAEL SMITH: Does New York have similar restrictions on the disclosure of criminal records held by DJS?

DAWN HERKENHAM: With criminal history record information?

MICHAEL SMITH: New York State does not have a specific statutory requirement for that. Under regulations, the Division of Criminal Justice Services have held criminal history record information confidential and will only release it if there is another statute in New York or a federal law that authorizes access to that information.

MICHAEL SMITH: In your experience do people who are employers in New York have any difficulty getting that information?

DAWN HERKENHAM: Employers?

MICHAEL SMITH: Yeah.

DAWN HERKENHAM: Well, I will have to tell you that I don't work for DCJS currently. I used to. Actually, I used to be the counsel.

MICHAEL SMITH: I remember.

DAWN HERKENHAM: You know what a hard time I gave you to get criminal history record information, anonymous information for research purposes.

MICHAEL SMITH: For research, right.

DAWN HERKENHAM: So you know I gave you a hard time on that.

MICHAEL SMITH: Right. I am not asking if you gave employers information. I know you didn't.

DAWN HERKENHAM: You know it. And I believe that that same strict confidentiality requirements are still handled and still used by that agency, and I will give you an example, for instance, and you may not agree with this. I know for a number of years, Big Brothers and Big Sisters have been trying to get access in New York State to criminal history record information, and because there is not a specific state statute that authorizes that, they are not permitted. So even something what some may consider a worthwhile endeavor, unless there is a specific statute that authorizes access for specific purpose, that information is not accessible.

MICHAEL SMITH: Well, all I can tell you is that I have seen a fair number of rap sheets that came out of the DCJS system in the hands of people who didn't have authority to get them and who weren't granted that authority pursuant to any procedure that you are describing.

PARTICIPANT: We know that is true. We know that is true. They appear in the New York Post.

MICHAEL SMITH: Yeah.

DARRELL SANDERS: But it seems access to rap sheets is quite a bit different than would be to this. I mean they are just much more available to thousands and thousands of people in the system, and this information wouldn't be. So that is a -- that may be a weak argument, too.

MICHAEL SMITH: But it's a weak argument except in the way that Phil is using it, that is the popular understanding of how good the protections are, the conventional kind you are describing, is the popular understanding is they are not very good. Even though you are all law abiding people, all right.

DOCTOR REILLY: Yeah, please understand I am actually a great proponent of this technology and its use, and what is motivating me is I'm trying to find the right path to make it most acceptable for good reasons to the American people, and we are among those few that have the opportunity to really explore the question in a meaningful way; and I think before we submit a final report, we should apply an acid test to ourselves and say, what would the most virulent critics of the system say about our findings and why. Do they hold up?

DOCTOR CROW: Should we hear from such virulent critics.

DOCTOR REILLY: Actually, I think we should. It's for both NRC1 and NRC2, we sent out our draft reports to people to have them criticized, much as papers are peer reviewed before.

CHRISTOPHER ASPLEN: Along those lines, I don't -- I don't want to affect the tenor of the conversation, because I think it's very valuable, but there is no expectation that this issue is going to be resolved today, or that any recommendations will come out of this group today, or that any consensus will come out of this group today. We do anticipate bringing in other representatives to come and talk to the commission from different perspectives. So this is just the beginning of the discussion on this particular issue.

DOCTOR CROW: Well, for a beginning it has been a pretty good discussion at least.

JUSTICE REINSTEIN: I just have a couple, a couple of conferences that I have been to, Doctor Chakraborty has said that with the 13 loci that we are at identity. So I guess I ask Paul and Dwight, Phil's question before us, why do you need to go beyond identity when you have got, you know, one in a trillion, or one beyond the world's population, or is that wrong in some instances?

PAUL FERRARA: No. I will give you a perfectly good example, Judge, something you can relate to. We had a data bank hit about a year ago. Because of the limited number of useful loci, among those 13, we developed a -- it was a likelihood ratio of 1 in 2,900. Now, this is a cold DNA data bank hit. As is typically the case, the prosecution and data bank hits had very little other evidence. The defense filed a Motion to Dismiss the case on the basis that one in 2,700 did not constitute reasonable --

MICHAEL SMITH: Sufficient evidence.

PAUL FERRARA: -- sufficient evidence, whatever the term is.

BARRY SCHECK: To do what, to resample the guy, or to --

PAUL FERRARA: No. No.

BARRY SCHECK: -- because that is not what you are going to present in court.

PAUL FERRARA: Yes, that was all we had in court, because absent additional genetic locations to look at, we were limited to only a few number of locations. So the Motion to Dismiss was granted. The guy is walking the streets right now.

DOCTOR REILLY: And that is the only time that has ever happened to you?

BARRY SCHECK: I don't even understand the example.

PAUL FERRARA: Thirteen is not always enough. So I want to go back and get that original sample, look at additional genetic locations. Now, what happens in that cold hit, we went back and got a fresh sample from the guy, all right, which is commonly the case. And we can then -- we then conducted additional testing on that, but we also went back to the original sample as well, at that time, to do the same testing on that material, but 13 loci, Judge, and eight in a straight sample is plenty.

BARRY SCHECK: You went back to the guy, and you resampled him?

PAUL FERRARA: Right.

BARRY SCHECK: So you didn't need the retained sample?

PAUL FERRARA: Right.

MICHAEL SMITH: Not to convict the guy. To convict the sample, you would have needed the retained sample.

PAUL FERRARA: Right, but what I am trying to answer is the question as to why 13 isn't necessarily always enough.

BARRY SCHECK: Let me give you the only example I can think of. I am trying to come up with examples to support your argument.

PAUL FERRARA: Okay.

BARRY SCHECK: And they are really out there. The only one I can think of is that you have a sample that is so small from the crime scene that you can only do one test, all right. I mean --

DOCTOR REILLY: That is not realistic.

PAUL FERRARA: No. The situation there was with the existing loci; and right now, I have got another case just like it, and it's an active case so I don't want to get into any details, but it's the same thing. It's less than 1 in 4,000 likelihood ratio, and we are scrambling around trying to look for additional locations. Now, I will admit that we can run more of those other genetic loci on a -- on this probable cause sample from that individual and not have to go back to the original sample, unless -- and this is I guess we haven't talked about it, and maybe my legal -- our legal arguments are really trivial, but what is the likelihood that future prosecution will be hamperedby the fact that the prosecution can't produce the sample upon which the original probable cause was based?

BARRY SCHECK: Assuming that you got the defendant, you took blood from him, you resampled him, and you got a match, zero.

DOCTOR REILLY: That raises a question that I would like to put out to those who know more about it than I do. I'm not even sure that I would characterize a blood sample drawn from a fellow at parole as evidence of any kind in a real case.

Is that really evidence in the traditional sense of using the word?

GEORGE CLARKE: It's probably something that as you tell a chronology to a jury, you would do in a very sanitized or safe fashion. It's like we do with fingerprints. When there is a fingerprint match from an automated search, you generally aren't allowed to let the jury know exactly how you found the person, and we sanitize it by saying sometimes, well, you have to give fingerprints to get certain licenses and so on to take basically the sting out of it. So as a practical matter that is not a real issue at trial.

DOCTOR CROW: I would like to raise one point I would have said yesterday, if I had found the time to do it, but one of the -- it's my fault for not having the time so that is all right. One of the things that the technical committee has wrestled with is that there is considerable sentiment in the group and especially on my part that it would be nice if we had sufficient accuracy in our testing that we don't have to worry about the possibility that there was a relative in the sample that you didn't know about, or that the population substructure were more extreme than the models that we make about that; and if so, then one might use the following criterion. We know that brothers or sisters, but I am especially interested in brothers, share one -- have a chance one in four of having identical genotypes at any particular locus irrespective of allele frequencies. So that one fourth is a very solid number. It's modified slightly by allele frequencies, which will differ from one group to another, but not enough to affect this number very much.

And we would sort of like to suggest that maybe if the criterion for evidence that is presented in court said not that the chance of two random people matching is such and such, but the chance of two brothers matching is such and such; and if they are any less related, it is less. I think that would be a very nice criteria, and it would get us rid -- it would get rid all at once of the necessity for typing people by race, keeping separate databases, et cetera.

PARTICIPANT: Is that one and four to the thirteenth then applying that thinking?

DOCTOR CROW: Roughly, except it's a little larger than one-fourth, because it modified the others. Now, 13 loci, the average match probability in this case is something like one in ten or a hundred thousand. It's not the high numbers that ordinarily appear in court. If you went to 16, 17, or 18 loci, then you would be reaching the same kind of numbers that one is customary to use with 13 loci with random populations. I don't know how important this is, but it's a consideration that I would want to bring up.

BARRY SCHECK: It wouldn't really affect this issue in the sense that if you wanted to present it to a jury that way, right, just get another -- just you can then do another four loci when you have caught the person and you want to prosecute, you can resample it. I mean in terms of doing database searches for all practice purposes --

DOCTOR CROW: I think you are right about that.

MICHAEL SMITH: This is an investigation.

DOCTOR CROW: I am really talking about what is presented in court.

BARRY SCHECK: That is different.

DOCTOR CROW: I just wanted an excuse to bring that up.

(Laughter.)

PAUL FERRARA: I would like to ask one question if -- with respect to the population databases and its use for statistical purposes, if we give the signal to the public that all these samples have to be destroyed; and as Phil said, yes, we could -- we did get statistical databases developed before convicted felon sample databases were developed, why -- what -- why would the persons -- why would persons feel comfortable giving us samples for the creation of these larger databases?

I mean aren't they going to be just as worried should not --

DOCTOR REILLY: Oh, you are talking about the creation of databases for allele frequency purposes?

PAUL FERRARA: Right.

DOCTOR REILLY: I see your point. You don't completely escape the argument, but I mean one way to do it is to have blood from a blood bank given to you in which there is absolutely no identifier, absolutely no identifier.

PAUL FERRARA: But, again, that is how these samples are stored, Phil.

DOCTOR CROW: There are two people I want to call on.

Did I forget about you, David?

DAVID COFFMAN: Yeah, I wanted to say a thing.

DOCTOR CROW: Let me call on you first then since I forgot about you.

DAVID COFFMAN: I am Dave Coffman, and I promise I will shut up after this, but basically someone mentioned why more than 13 STR loci. We are trying to, as far as having actually been someone with databasing experience in this room, we have had technology changes. I can't tell you what the new technology is going to be. I can tell you right now there is rumblings in the community that they really wish that we had included Y chromosome as part of the core loci already. I mean we just set these 13 back in the fall of '97, I believe it was, and there is already talk that we need Y chromosome issues. It is not -- it's not the fact that 13 does not give you a huge number in a pristine sample, but the problem is the more sensitive we get, we are going to be picking up mixtures more often, and sometimes they can't be resolved, and your numbers plummet at that point.

Also there is talk about automating, using automated procedures for looking at sexual assault kits, doing the differential extraction. If we have Y chromosome issues, that would greatly enhance that capability. So there is just -- I mean this is just what I -- you know, I don't think on my feet very well, but that is just what I came with now, and that is something that we need to consider.

And also as far as my being in civil service, we always have to think of the cost-effectiveness. Now, I know when it's public safety or whatever, we shouldn't think of that, but the reality is we have LA with three people doing the sexual assaults -- not doing sexual -- testing sexual assaults in LA. Sorry about that. You know, the fact is it's hard to get money for civil service. Crime labs have never been staffed properly. You know, they didn't put 100,000 crime lab analysts on the street a few years ago. They put 100,000 cops.

(Laughter.)

DAVID COFFMAN: You know, so we are always understaffed. We are always underfunded. We always have buildings that even if we had staff and funds, we couldn't squeeze them all in there. So that is an issue we have to consider, but my biggest plea is the effectiveness of the database. If you start saying that ten years is somehow an adequate time that a technology can be in place, we have already had that, and it took us seven or eight years into the database formation to really start becoming effective. And to say that we can only account for from here on out, we can only have two years out of ten that were any good is just shortsighted, and that is just all I want to say.

BARRY SCHECK: If the state legislature turned to you and said, I will give 50 analysts, right, because minority caucuses that have special fears about the collection of blood, right, that is a bunch of votes, even in Florida, right? And this group and that group say I will give you 50 analysts, which is how many million? How many million would 50 analysts, more analysts be for you in the State of Florida?

DAVID COFFMAN: It would be quite a bit.

BARRY SCHECK: Okay. If you destroy the blood, you would take that deal in a heartbeat.

DAVID COFFMAN: I got -- we got 69 new analysts in the State of Florida keeping the blood so...

DOCTOR REILLY: But that is not typical of the nation.

DAVID COFFMAN: We show effectiveness.

BARRY SCHECK: What I am saying is when you start putting in this balance, you know, and you start looking realistically at the politics and how people really feel about the government, you know, trust us, we are the government. We don't do anything with your DNA. Right. But if you destroy it, you can get money; and so far, you know, I don't see a lot of success, do you, in terms of money for this?

DOCTOR CROW: Let me call first on chief and then on Jim Wooley.

TERRY GAINER: Thanks. I actually think I want to go in the same direction David did and talk a little about Dwight. Again, we are spending a lot of time talking on the court aspect of this, which is clearly very important on the constitutional aspects, which we spend a lot of time before on the privacy issues, but it seems to me that we are losing the focus somewhat of the investigative value of a database. Again, it seems to -- we have picked the American approach versus the United Kingdom. And however the commission ultimately decides this, we ought to have our arguments in somewhat alternatives, so that if one is going to destroy the blood bank, then this is the social criminal victimization impacts of that; and if -- and as you mentioned maybe that we mandate you keep STRs or have the technology committee get gateways for us that we have heard are always going to come in so many of these technology areas, but never seem to do, because if we only focus on the fact that, again, that it is a privacy issue, or if we argue, you know, the cute turn of the phrase "trust us, we are the government," then we do lose the potential to prevent victimization and to solve crimes, and I think we ought to devote just a tad more time to that and of the argument versus where we have spent an awful lot of time thus far.

DOCTOR CROW: Jim. I saw you too late. Sorry.

AARON KENNARD: Well, I'm glad you finally heard from one of the other law enforcement there. There are four of us on this panel, and you have not heard from any of us here, and I suggest we be very careful if you are going to be bringing on some of these minor or so-called minor groups that have special interests, because what of the overall community do they represent. I am part of the government. I am very offended when you suggest that we can't ever be trusted, and yet you can get a fingerprint card, you can get a crime scene analysis, and you can get a history of -- a criminal history of anybody you want, if you have got enough money, or if you have got enough contacts through the press. So I would ask this commission not be guided that you -- I think possibly listen to some of these people that it may be of some value, but they are not going to be directing or guiding us in our efforts. Be very cautious in some of these special interest groups. What percentage of society do they represent? I deal with the ACLU on a constant basis. I get a lawsuit every week from the ACLU, but the general public in Salt LakeCounty are very offended by these people.

DOCTOR CROW: Jim Wooley.

JIM WOOLEY: I wasn't in response to what Terry said so I will wait until people want to follow that up. I just had a couple of questions for Paul and his group actually that was unrelated to what Terry said.

JEFFREY THOMA: Just a really quick point. Terry, I don't think we lose anything with regard to matching up to the database by destroying the samples. Absolutely nothing.

TERRY GAINER: It's the technology changes, Jeff. That is what I understood we would lose. We would have to begin recreating -- excuse me. Just so I understand so we can speak the common language. I guess I have understood from a layman's perspective that if the technology changes, then we would have to -- have to have either new samples or create or start or resample what we had held onto or begin the creation of a whole new database. Now if we get gateways, then I guess I don't have a problem with that.

JEFFREY THOMA: You either run them in parallel, or you have to keep the STRs, just as I said, you have to keep the RFLP and the PCR D.Q. Alpha, whatever you have used previously throughout a certain time frame anyway.

TERRY GAINER: But I do think that is a cost, which was not going to be borne, or we just have to discuss the cost of that.

JEFFREY THOMA: And I think the cost is less really of doing it parallely and rerunning it in those isolated incidents when you need it, if the new technology advances.

Then the storage, then the other issues of cost that are going to occur from keeping the whole block, I think it is something you have to balance even within your costs, even by doing it that way, and you are not even taking into account those privacy concerns which trouble me greatly, too. But I don't think you are hurting law enforcement by doing it. I would not want to do that. I think, obviously, we have to figure out a way to do it.

TERRY GAINER: Just so -- I am missing something, and then I will need the technologist to tell me, because again it seems to me we would be losing something. If we change to technology number three, assuming RFLP and STR 1 and 2, I guess I am understanding, absent some type of segue back into the STRs, which will be much, much larger numbers than we had to deal with the conversion of RFLPs to STRs, that when I want to go search a data bank, I won't be able to do it, because the majority of the offenders will be held in a data bank for which the current technology won't let me back in there. That is the way I understood the problem.

JEFFREY THOMA: Otherwise, if you want to do that then you are going to have to, if you do get the new technology, you are going to have to rerun everything if you want to do it that way, and that is incredible, the cost of that would be -- Paul, you know that. It's impossible.

PAUL FERRARA: Well, no, no. It's not impossible.

JEFFREY THOMA: Anything is possible, but it would be extraordinary.

PAUL FERRARA: No. No. Look at the RFLP to STR transition. In Virginia, and Phil is right, in NRC 1, we specifically said that we didn't recommend the creation of large RFLP databases, because in 1990, when we wrote that report, we knew that STRs were on the horizon. So we said that the development of large RFLP data banks wasn't wise. And we were right, because over the next seven years, in Virginia, we had run some 15,000 RFLPs. Now, we don't bar RFLP pretty quickly, and advisedly so, but there are a lot of laboratories in the United States right now who are still running RFLP analyses.

Now, when they get set up to run STRs and get the funding, they will be able to go back like we did and rerun -- and economically rerun their -- they will have to rerun those RFLPs; otherwise, they are going to have to start from scratch again.

Now, let's say ten years from now I can run -- I can run samples at one-tenth the cost of STRs. Currently, let's say 50 bucks a sample. Let's say I can run them for five bucks a sample, and they are equally as -- as equally as reliable, specific, sensitive, as the STRs, and we want to go to all -- to say arrestee testing, or other states choose to expand, as I wish they would, their statutes to all felons. There is going to be quite an advantage to wanting to say, well, let's jump to that new technology, but that is going to mean rerunning now all the STRs, if you have made that transition, but I can see that -- you know, I can see that happening.

TERRY GAINER: But from the integrity of the database, I mean if the database has a value, that is what we are going to have to do, or we are starting all over again, and I think that is the argument I hear from David and Dwight and others, and from just a practical crime solving, crime prevention, that is a value -- that is a bigger question as anything that we are dealing with here.

DOCTOR REILLY: Although I must say, I haven't heard an answer as to why, if there does emerge a new technology, which is almost certainly too powerful and cost-effective, systems couldn't be run in parallel on a sample. I have not heard an argument.

PAUL FERRARA: Well, you can't, because whatever sample of technology that you are running on crime scene samples, that has to be the technology that your database is based on.

You can't search a -- if you run an STR, I can't search an RFLP database.

DOCTOR ADAMS: Of course not, but if you had -- let's just say that by the time Y sequence technology emerges on the chip, you got two million samples stored in databases in the United States. You really don't -- I mean you really don't want to abandoned STR technology at that point, do you?

TERRY GAINER: I don't know that we would in that area, Dwight, but I can say what is going on in some other technology where I haven't seen the compatibility, and that is in the whole weapons issues in new technology, where one group of people has us doing IBIS, and another has us doing brass catcher, and another has us doing drug fire, with the promise over the last three or four years that somehow these would all be compatible, and they are not.

So, consequently, you are asking departments and labs to do something that is very cost prohibitive, and as a consequence, not inner solving our cases, and again, locking people up and getting the right ones off the street.

BARRY SCHECK: It happened here initially with RFLP D.Q. Alpha, but now we are on the same system, but just so we are all clear about the limited issues, let's say that we decide that we want to do, let's say for the sake of argument, mitochondrial testing, all right, on everybody, because in terms of trace evidence, it's such a good and sensitive identifier. Well, the way you would start is that, because you are absolutely right, you have to do the testing initially on the crime scene samples. As new people go into the system and are convicted for designated offenses, you begin to type them as they come in, and the statutes would say, as they go out you begin to type them, right? And then the pool of people that -- would you go back and try to resample the ones we don't have yet anyhow, right, the people on supervised release? I mean, you know, you have to start thinking this in terms of system. When the new technology comes in, as long as you are typing the new people and the new crime samples and the people coming out, you are really going to capture for the most part, the people of greatest interest; and eventually over time as the technologies run, you bring them in. So you are not necessarily -- it's not such a, you know, huge thing that you have thrown away the initial samples for the STRs.

TERRY GAINER: That is a good point. Again, when we do these alternative arguments, let's think about the alternative solutions and the alternative cost, and that cost just in dollars.

JEFFREY THOMA: Oh, I think we agree completely on that, Terry.

DOCTOR CROW: Jim Wooley, you have been quiet for a long time.

JIM WOOLEY: Is it accurate that the discussion seems to suggest this to be the case is that your reasons for retaining the samples in the group are those listed in order of descending importance?

PAUL FERRARA: No. No. No. No. No. And it's not meant to be all inclusive. It's just what we could think to put together.

JIM WOOLEY: I mean I know a lot of the folks in your group, and I know you have put a lot of thought into it, but it seems like that is where the discussion is currently framed. I mean it seems the -- the idea that we need to keep these things, because as number three suggests, and you talked about that, that we bungle handling of samples for database reasons is something that seems to me, unless you really have a lot of that going on, that seems to me -- I'm kind of uncomfortable even saying, I mean, do you really have -- do we really have these situations here?

PAUL FERRARA: Yeah, sure.

JIM WOOLEY: Do we have cases -- have you been involved where you get a hit, and you go back, and you find out well that guy was in jail, or he couldn't have done it. So then you rerun it on stored DNA and then somehow find that some mistake was made that saves you now going to get a sample from that guy?

PAUL FERRARA: No.

JIM WOOLEY: Because if it doesn't happen, maybe we shouldn't be asserting that as a reason for --

PAUL FERRARA: I think -- no, that hasn't happened, Jim, but it could, because when you are -- realistically, when you are handling, as we will be, if not already, tens of thousands of samples, transporting them, excisioning them, taking them all the way through the process, I would be a fool if I sat here and said there is no chance that mix-ups will occur.

Now, maybe what you are saying is what is the worst-case scenario if there is a mix-up, further investigation when you go to get probable cause to go arrest this guy, you are going to get a known sample, you are going to check it against your crime scene sample, and you are going to say, hey, wait a minute, this doesn't -- this isn't the same profile. What happened? Well, meanwhile so you say to this guy, or whoever you have arrested, oops.

JIM WOOLEY: Sorry we arrested you; and if we had had your stored DNA, or we had had stored DNA, we might not have had to have done that, because we could have --

PAUL FERRARA: We could have checked it before that officer went out and arrested that guy.

JIM WOOLEY: My only point is that that seems to be sort of, in terms of anecdotal support for there being that real problem, that seems to be kind of small. It seems that if you just list them all one after another, it seems like the technological change argument. That is the real one, and that is the one where this discussion needs to take place.

PAUL FERRARA: Clearly. Clearly.

JIM WOOLEY: And the idea that we need to do this, because people are bungling database samples in handling them, I think, is where the discussion doesn't -- but the technological change area, when you look at history, I mean there is reasons why that really needs to be flushed out just on those issues.

PAUL FERRARA: I mean, again, you are absolutely right. I mean there is no -- I mean clearly the main reason we don't want to do this is protection of our samples against some future technology.

JIM WOOLEY: Right.

PAUL FERRARA: But we wanted to point out there are a myriad of reasons why we in Virginia go back to those database samples on a routine basis for a variety of reasons. And so then we looked at all the reasons why it's nice to have those samples, and then we said, all right. Now, where are the privacy and where is the risk of these samples falling in information being released? And we, from a laboratory standpoint say, there ain't any. So what is the point of taking a chance on destroying the samples and eliminating a variety of advantages to keeping them simply for the sake of giving, giving the warm and fuzzies to the general public or the ACLU --

JIM WOOLEY: Yeah, I understand that.

PAUL FERRARA: -- when, in fact, what people ought to be looking at and be concerned about, and it goes beyond the purview of this commission, is the -- is the willy-nilly nature in which samples are collected all over the country, but...

DOCTOR CROW: Let me. We haven't heard -- let me say it another way around. I said at the beginning that we would give people, who are attending the meeting here, a chance to speak after each presentation. I think this is a good time to do it.

So let me ask if there are people that are around the rim of the room who would like to say something. Yes, you first.

DETECTIVE FOSTER: Hi. I am Detective Foster from the LAPD, and I wanted to present kind of a different -- you are all scientists and very knowledgeable experts in everything, but the warm and fuzzies, for god's sake don't ever say that outside of this room, because we want the warm and fuzzies. There are very few people who do vote, our Vietnam era raised on, and all the Pentagon papers stop, and General Westmoreland and all the lives and all the people that died because of that, that is who we were raised on.

And look at the syphilis experiments. Yes, they are one in a million, but they are there, and they are in the American memory. You can't ignore that. You are scientists, but you are not the cops; and if your entire case is focusing on one drop of blood, then that copper hasn't done his job, or that guy is such a good criminal that he is going to get away with this one. Everything cannot be based on that one-in-a-million shot. I am not a scientist, but you can't -- you can't tell me yes, just trust the government, because, no. Yes, we should trust the government, but always without being completely naive and without being completely ingenious about the whole thing.

Mr. Scheck is completely right, and the gentleman here whose name -- I can't see, you have to convince the American public, and I admit, this is it. We are under 40, and we are young, and we don't know what the hell is going on, and you just can't tell us that this is a really good idea, because maybe if we can't -- we need more than the 13 loci. Then that is just not a good argument, because we don't know what the devil you are talking about.

So when you make these policy decisions, positions, and especially as a police officer, if you have got a 71 percent recidivism rate in the county of Los Angeles, then why are you worried about your data bank. They are coming back. They are coming back within one year. You have got them coming and going; and next year, if you can't get them this time, then you get them again. That is seven out of ten of all the prisoners that go in LA county alone. If you guys need more comparison samples, you are welcome to them. Come to California.

Thank you.

DOCTOR CROW: Thank you.

Anyone else? Please.

PARTICIPANT: What was her point?

DAN ERLICH: Hi. I am Dan Erlich. I direct the Whitehead Institute Chip DNA Project. I sort of see this group struggling with trying to bring this very powerful technology to bear for the public good, and it has taken us a long time to even come up with the 13 loci to have enough of a mandate to get started on this, and I have seen the struggle of getting going as being something which we shouldn't -- we can't afford to lose momentum about; and at the same time, we have to acknowledge that the technology will be changing in ways that have been brought out. So what is certain to happen is that the costs of doing the assay will plummet. It will -- it will -- in fact, with current technology plummet if the facilities were properly centralized. The point where they are, for example, in genomics for the Human Genome Project. Immediately, if you were to apply it by economies of scale, you would find that the assay cost would go to a small fraction of what it is. As the new technologies develop, it will further plummet. And it's also certain that the consensus that we have got to achieve the 13 loci, that will prove to be an incorrect decision that we will find that in the future there will be more powerful ways to apply similar technology, which will be more effective in the public good of crime solving.

And because the assay cost will plummet, and because the choices in what went into the database will change, we will want to redo the database, and what that will mean is the cost will be entirely associated with picking up samples. It will, particularly when these people are no longer available, and you have to send somebody out to collect it. So this is, I think, the best reason for keeping the samples is that you are going to have a far more effective tool if you are able to modify it over the next period in time.

DOCTOR CROW: Thank you.

We perhaps haven't emphasized, as we should, on the fact that the changes are likely to be instrumental in modifications rather than substitution of total new things.

Yes, Phil.

DOCTOR REILLY: I would like to put a question to both Doctor Erlich and Doctor Crow. Looking at it from the point of view of population genetics and the power of the 13 STR loci and admitting full well that there are better technologies ahead, if we froze this technology in time and said we are not going to go with a different set of loci, whatever, but we will try toimplement technologies that reduce the cost of asking questions, would that be such a terribly bad solution over the next decade?

DOCTOR CROW: That is going to happen anyway.

DOCTOR REILLY: It's going to happen anyway. You said that, you know, in retrospect, it would look like a -- you said incorrect when you referred to the 13 loci.

You really didn't mean that, did you? It was quite correct at that moment in time, and it's a damn good system.

DAN ERLICH: In fact, the decision needed to be made two years prior, but -- and also it is true that it will be highly desirable to change it. So I think that what will happen is that the method of assay will begin to affect the choices as well, because we will find that we, if we had only chosen a slightly different 13 loci, the cost could be reduced by another factor of four, because we will be analyzing shorter strands of DNA, or something like this. So I think it is certain that the database, the choices will need to -- the best choices will change with time in fairly short time horizons.

DOCTOR CROW: I don't have anything to add to that. I'm just going to preside.

Let me call on Chris though for one other point here. Just give me time for an intermission, and he wants to make one announcement first.

CHRISTOPHER ASPLEN: We were fortunate enough to have the bureau provide us with these beige documents for state DNA database statutes. We don't have enough for everyone; however, if you contact Dwight, we will make sure that anybody who doesn't have one and who wants one will get one. This is primarily the result of Dawn Herkenham's enormous efforts on the weekends and other times to put this together, and it has been great value to myself as commission staff already, and I think it will be a great value to the commission's work. So we appreciate the FBI providing that to us.

DOCTOR CROW: My watch tells me it is time for us to take a coffee break. Let's promptly reassemble in here at eleven o'clock and hear Phil Reilly.

(There was a short break taken.)

DOCTOR CROW: Unfortunately in my tunnel vision this morning, I overlooked a person who was very eager to say something. He is still here so let's have your word in here, and we then will move on to Phil.

JOE BELLARO: Just once again, I am Joe Bellaro. I am the senior criminalist in the Boston Police Crime Lab. I want to throw in a couple of things. I have forgotten much of it, but a couple of things.

Just to echo Doctor Reilly's points, I have been in this community since PCR has been in this community, and I seen it go from snips, D.Q. Alpha and Polymarker work, so snip is nothing new to this community, but I have seen it cycle. Then we have gone to STRs, and now there are thoughts of maybe going back to snips eventually. But just one thing that he said that I agree with is the jump from RFLP to PCR is a significant jump, but any jump since then have really been minuscule, it really has, in terms of the background of jump to PCR. So I think it is possible for the technology to, at this point, plateau and maybe fine tune and add, but I really don't think it's going to change dramatically.

In line with that, sometimes the user doesn't have absolute control over the technology, also, and so manufacturers and industry and academia also dictate certain things in the community, which, you know, we may have to deal with, and that may require resampling some or all samples.

Then the other point just has to do with privacy, and these are really interesting discussions to me, but I just did want to mention this. The privacy issue, I think, really can vary depending on who you poll, and I think this room represents upper middle class, educated America, which has its perception about issues of privacy, but I think if you get in your cars, and you drive to Roxbury in Boston, which is a lower socioeconomic area, and you ask the people that are the victims of most of the crimes and are suspects or perpetrators of the crimes, I think you will get a vastly different opinion as to should samples be destroyed, should they be kept, or should they even be collected at all. And I know you have considered that. These are a little bit new to me, so I am just trying to catch up, but I think it's important maybe to think about that end of the area, also. That is a big concern.

DOCTOR CROW: Well, thank you. Thank you.

Previous	Contents	Next
Back to National Commission Main Page