General Question

robmandu's avatar

How accurate is DNA testing really?

Asked by robmandu (21306points) July 22nd, 2008

Just shooting from the hip on this one, folks. No real research done by myself here. Just read this article from the L.A. Times and it made me wonder.

You know, when they say such-and-such a match has a statistical probability of 1 in a 113 billion, what does that really mean?

It doesn’t mean impossible. Just improbable. And if the improbability calculation is built on faulty premise, then the number means nothing at all.

Should DNA testing be pulled down from its lofty pedestal in courts of law? Right now, it seems kinda incontrovertible:
“We have DNA evidence that puts you at the scene, Mr. Mandu.”

“No, but wait, I wasn’t there. And it’s just a statistical probability you’re using.”

“Yah, try that line of reasoning on the jury. Buwahahaha!”
Increasing the number of required locus points seems to help make the statistical probability more palatable. But does that really solve the problem, or just tackle one of the symptoms?

There’s no perfect crime-busting tool. No silver bullet that only kills the criminal. Can coincidence only ever be limited, not eliminated?

Observing members: 0 Composing members: 0

8 Answers

Harp's avatar

Yep, could be your twin.

nikipedia's avatar

I only skimmed the article but it sounds like they’re talking about the good old multiple testing problem. If you have a 1% chance of something happening by coincidence, and you do it 1 time, then it’s probably not a coincidence. But if you do it 100 times, is coincidence still “improbable”?

If you compare a DNA sample to 360,000 other ones, it would be statistically improbable that you would not get some matches…I think.

Lightlyseared's avatar

That is why law is an art not a science.

sks485's avatar

DNA Testing is extremely accurate. I believed it is 99.99% accurate. DNA Testing chances of being wrong are remote and almost beyond possibility. They would not be freeing people in jail for wrong crimes if it was not this accurate.

robmandu's avatar

@sks, it’s even more accurate than 99.99%, more like 99.99999999999% accurate (supposedly, but that’s what the article argues about).

But let’s use your number, just for argument’s sake. In the U.S., there are 300 million people approximately. If you were using a technique to identify an individual with 99.99% accuracy, that would mean that 3,000 people in the U.S. alone might match as false positives. Distributed evenly to each state, there could be 60 people that might’ve had reasonable geographic access to commit the crime you were put away for. (Expanded to the whole world, that’d be 65 million false positives!)

If I had to do it over again, I very well might try to take more than just Stat 101 in college. Understanding how probabilities can be (mis-)used seems to have an increasing importance in life these days.

Lightlyseared's avatar

OK, having bothered to read the article now.

Everyones DNA is different. If you were to sequence the DNA of everyone in the US no two people (excluding identical twins) would have identical DNA. However it is important to understand that certain parts of the genome dont really change that much. For example the bit that codes for haemoglobin doesn’t vary much or it wouldn’t be able to produce a molecule that would be useful and the individual would die. Only about 3% of the human genome actually codes for anything useful the rest is described as junk DNA and it is this part of the genome that is used to differentiate people. That is why when the researcher in the article found two people with similar profiles they did not look a like as the similarities were in the junk.

As for the accuracy of DNA testing it has to be recognised that there are problems with the techniques. Crime labs don’t sequence the whole of the piece of DNA as that would take way too long so they only look at certain sections of the whole genome. Over the years the accuracy of the testing has improved but that does not mean there is not space to improve in the future.

Until it is possible to sequence each sample entirely quickly and cheaply there will always be room for mistakes but that shouldn’t undermine the technique in law enforcement. There is very little scientific basis for fingerprinting for example yet it is still used daily without being questioned.

sks485's avatar

DNA may be wrong in a few certain instances. However, it has freed many innocent people from jail. It is not perfect but with in 10 years the accuracy will be even more than it is today. Techniques will also be much improved.

Response moderated (Spam)

Answer this question




to answer.

This question is in the General Section. Responses must be helpful and on-topic.

Your answer will be saved while you login or join.

Have a question? Ask Fluther!

What do you know more about?
Knowledge Networking @ Fluther