I think it would be great for everyone's DNA to be collected and available for researchers, police, demographers and others to use. This could be collected just after birth and registered.

I think the best way would be to have a universally unique identifer on each DNA record or records, then if the police did a search for a suspect and got some hits in the database, they would have to get a court order or similar to retrieve the actualy Social Security number, real name etc., that this identifer was linked to.

That should solve all ethical concerns.

Because handing governments that much information and power has never gone awry. Nope, perfectly safe, we can trust them to not misuse the information. The governments of the world have never abused their authority before, if you can't trust them who can you trust?

There's also the practical ethical concern that trying to enforce any such universal scheme in the US would surely result in some arrests and deaths, as full compliance would not be voluntary. Of course proponents of such a scheme would try to lay the blame for any such problems on the unreasonableness of the resistors, but I'm less sympathetic to the authoritarian side.

http://news.yahoo.com/s/ap/20060...TdmBHNlYwM3NTM-

There is always a false-positive (match) rate. It compounds poorly with people's strong tendency to focus on the most obvious evidence. Given a DNA database and modern PCR amplification techniques, police will try to collect DNA and find a match. They will neglect other forms of evidence, assuming they are less probative and therefore unnecessary. The police will then arrest any matchee, and when he is prosecuted the only evidence available (due to police insouciance) will be the DNA match. If the matchee is not, in fact, guilty (false-positive) he will be left with the unenviable task of proving a negative ("I didn't do it! I always chew on pens and I think I left that one on a bus by accident! I don't know how it got into the girl's car!")

This problem can only get worse with a bigger database. It's not that taking DNA only from those arrested would prevent false-positive injustices--but that taking DNA from everyone will inevitably increase false-positive injustices.

The problem is worst with "trace" evidence (that is, tiny amounts of DNA swabbed off of random object and assumed to have been shed by a visitor to the scene--or anything other than semen or visible blood evidence), because our ability to detect DNA now extends to detecting amounts small enough to wander anywhere. Step on someone else's sidewalk-spit and you may leave traces of his DNA everywhere you walk for hours.

I think we need an evidentiary rule that "trace" DNA evidence can only be used to corroborate other evidence--and may not be introduced at all without some threshold showing of other (often circumstantial) evidence.

If a national DNA database is something you'd prefer not to occur, then now is the time to come up with the best argument against it. The database framework already exists, and rules for inclusion are becomming broader with time. Vague dystopian fears will not be enough to argue against their utility in criminal investigations.

Mark Seecof wrote This problem can only get worse with a bigger database. Actually, the problem will become easier to quantify (and thus control) with a bigger database. False positives are due to genotyping errors. The true rate of errors will be detectable in the rate of obviously false matches. Finding the DNA of babies or grandmothers at a crime scene, for example, will make clear (to the public, at least) the limitations of the technology.

The most worrying aspect to me is the identification of genes that have possitive correlation to criminality and the persecution, or even elimination of it's innocent carriers in some attempt to create a "perfect" society. These correlations could be due to other things like some "resistance to oppression" or "freedom loving" genes causing collateral anti-social behavior, but also being a net benefit and survival ensuring to society. And even more importantly, myself and I'm sure many others would be ethically opposed to descriminating against people for the genes they carry.

And the proposed solution of making it universal doesn't alleviate this concern, just because the database contains records for everybody it does not preclude the government from later comparing the identities of convicted criminals to this database, or comparing a list off all records that were subpoenaed to courts and led to convictions back into this universal database. I fail to this how making it universal allays these concerns at all, except for adding one bureaucratic step.

The best arguments against a database of this type are the same as the best arguments against a national ID card, domestic spying, keeping of your phone records, or just wandering into your house to see if you're doing anything illegal without a warrant.

Their utility in criminal investigations is inconsequential so far as being required to submit DNA samples absent of any criminal charge violates at least your fourth and fifth amendment rights. Fourth because it is an unreasonable search (warantless, involuntary), fifth because it undermines due process and, I'd argue, forces one to testify against oneself. It further violates the, admittedly, court-interpreted right of privacy.

There's little evidence that government can be trusted with any power granted to it: we've seen this over and over through history, and even the best governments go power mad (emminent domain abuse, the War on Drugs, McCain-Feingold) given enough time and resources. Just handing them what amounts to a pile of free evidence is a monumentally bad idea. Further, one has a right to be presumed innocent, and a DNA database says "we presume you'll commit a crime, so we need to have your DNA in case you do".

Timothy, I'm sympathetic to your appeal to privacy, but my casual reading on this topic over the last few days suggests to me that courts will not see it that way. DNA can be collected from a person by touching their skin with a piece of sticky tape, thus reducing concerns about the method of collection being a physical violation. Moreover, we're constantly shedding DNA into our environment, which reduces the expectation of privacy. A lot of these things have already been decided wrt fingerprints, and so moving to DNA will be an easy step.

I can imagine that sample collection won't be made per se involuntary, but rather it will be made a requirement for getting a driver's license or something like that.

As you can tell, I'm not optimistic about avoiding such a future. Given that, I'd prefer to have these decisions actually debated.

Rikurzhen: "False positives are due to genotyping errors."

Sometimes. They may also result from error in sampling, labelling, and processing; data-entry or storage; etc.

When you collect a "trace" DNA sample, there's some probability that it bears no meaningful connection to any crime committed in the vicinity. If the sample is actually irrelevant, then even a quite valid match of the sample to a DNA database entry results in an investigative error, a false positive indication of connection between the investigative goal and the DNA database subject.

Rikurzhen: "the problem will become easier to quantify (and thus control) with a bigger database."

The "problem" to which I referred was not that of laboratory DNA matching error (which in any case would be easier to control with a smaller database), but rather the problem of police laziness.

Given a DNA database covering more people, the police will make less effort to discover evidence that could point to people unknown to the database. Given a "universal" DNA database, the police will neglect virtually all non-DNA evidence.

It would be nearly impossible for innocent investigative subjects identified by DNA to acquire evidence neglected by the police. Neither they nor their agents will have timely (or perhaps any) access to crime scenes or witnesses.

As for people becoming disenchanted with a high error rate; well, how will they ever learn about it? Especially when the police gather DNA, look for database matches, discard some false matchees on discretionary grounds ("no white guy who drives a Lexus could have been there--the neighbors would have noticed"), then prosecute the remaining matchees on "DNA match" alone, with the burden of proof reversed ("what, you have no alibi? Your DNA was found at the scene--you must be guilty!").

Sometimes we talk about stories like those of police looking for a rapist who left a semen sample behind. In that kind of situation, we have more than "trace" DNA, and just from where we found it we think it likely has to do with the alleged crime.

We have to consider other scenarios, though. DNA extracted from dandruff found on the victim's clothes is likely less probative than a semen sample, yet police will be matching that sort of sample much more often. What if a matchee had merely danced with the victim in a nightclub a week before she was murdered--by someone else?

Liv,

Thank you for the link to the RNA inheritance article. Very interesting.

Mark, if the police start getting hits for people who obviously didn't commit a crime, then the failures of the system will be laid bare. If the only people in the database are those who plausibly could have committed a crime, then these false positives cannot be as easily detected. If there's a nontrival jeopardy of false matches, then best that we all face that jeopardy than just a subset of the population -- so that we'll all be inclined to do something about it.

This time, i'm going to try my very hardest to contribute something *relevant, considered* & sincere to said discussion, rather than babbling on about myself or my deficiencies.

My take on this issue is essentially what, i believe, timothy, above was getting at.

I believe that governments cannot be trusted, period. *Why* do i believe such? Well, because governments, especially modren ones, don't exactly, ahem, have reputable history w.r.t considerations for citizens.

What i'm saying is that government is a somewhat unstable entity.
In other words, you never really know when authoritarianism or totalitarian tyranny could gain hold. And then, what do you know; all the previously safe & useful policies & implementations could be turned *against* the people, or against some people, at least.

Well, i tried to be sincere or relevant here, so if i get deleted here, i'm not sure why. I've only expressed my views, which i don't believe to be unusually uninformed or ignorant in that regard.:)

Crime probably isn't a big enough problem to justify such a system. At least in the places I'm likely to live. Why are you so eager? You're perhaps more interested in the science potential than in stopping crime?

A worldwide database seems inevitable to me, although that may be measured in centuries, not decades.

Many criminals aren't stupid, and we can expect to see DNA routinely left at crime scenes, planted by the actual perpetrator.

Outed! Rikurzen is actually

Hmmm, link got dropped somehow. Post was supposed to conclude with "...NYC Mayor Bloomeberg" and the following link:
http://www.breitbart.com/news/20.../ D8HQE6B80.html