The Dark Side of DNA

Not to be paranoid, but one thing the article didn't touch on (unless I missed it) was the enormous potential for outright framing by police.

It's awfully easy to get someone's DNA, often without their knowledge. And law enforcement is keenly aware that DNA evidence presents a "slam dunk" in court. It seems plausible that more than one unscrupulous officer or DA could and would use planted DNA to boost their reputation and rid their district of people they think of as "undesirable".

I'm all for punishing the guilty. But given collection errors, lab errors and fraud, maybe we should consider restricting the weight DNA alone can carry in pursuing a conviction.

Too bad for Teresa Halbach that DNA set Steven Avery free on another murder rap

When I think of how confident we used to be, back in the day - you get some test done by your doctor, and you believed the results without ever thinking twice. Nowadays, errors in medical testing are the scandal that's always fresh in the news.

I find it hard to understand that the people running these DNA labs do not have strict measures in place to prevent contamination errors. And what kind of training did the lab techs get? It is outrageous.

DNA evidence and some other scientific measures are usually presented in "percent certainty" when it comes to court, so the margin of doubt is very clear. But the history of errors in the lab also reduces percent certainty. They should be checking and triple checking every sample and keeping a log of the errors discovered - that should be factored into the degree of uncertainty.

The "163 trillion to one odds" should have sent up a red flag as a bogus calculation. Victim and defendant DNA should not be tested in the same lab; instead test them in well isolated locations and compare the printouts. Even on "six degrees of separation" grounds, let alone cross-contamination possibilities, the chances of some association between samples would be way higher than those odds.

and the fact that there is only a 162 trillion to one odds that you exist!

At least not as misconceived as the Hoyle-Wikramasinghe 10^40,000 odds against evolutionary emergence of life.