Slashdot Mirror
DNA Extraction From Fingerprints
Myriad writes "A Canadian scientist has developed a new way of gathering DNA evidence for analysis using fingerprints. The new test can extract DNA in 15 minutes - even from a print stored for many years and in varying conditions. The patented extraction technique consistently produces ~10 nanograms of DNA. Analysis generally requires 5-10 nanograms, although it is possible with as little as 0.1 nanogram."
...as I think it will seriously make finding the guilty easier, and seperating the innocent from the guilty. If I'm not mistaken, it currently requires a judge's order in the united states to collect a DNA sample. Now all you have to do is dust something touched by the suspect to get a DNA sample.
Edward@Tomato - /home/Edward/ man woman
man: no entry for woman in the manual.
"Qua!?"
You always have the problem of getting it admitted in a US court. Expect big battles over this. For example, if the judge isn't the most tech-savvy judge around, someone could bullshit him/her into believing that the DNA samples are unreliable. Also, you have the BigBrother concern. However, fingerprints are already considered uniquely identifying. The only added problem w/ getting DNA from fingerprints is technology that is still years away, if it even comes to the market -- predicting people's characteristics/future death from DNA. However, the only reason to get the DNA from the fingerprint and not from the sweat that the person left, or the hairs that the person left, etc., is because of the storage factor. So, while people worry about their DNA being extracted from a fingerprint, they should be more worried about all the hair follicles and skin cells they are leaving behind that also give away their DNA.
If you already have the fingerprint, why do you need the DNA? Most criminals (or at least those arrested and brought to jail) are processed via fingerprints and that is what's stored.
/me puts on his tin foil gloves
Does this indicate a move toward DNA databases instead of fingerprint ones?
Will this save any time or effort on the part of law enforcement agencies?
Will newborns have their DNA sampled shortly after birth?
MMORPG fan-boy? Prove your worth
fingerprints are already each unique why do we need to get dna out of them?
This could possibly lead to more false positives than now. Say you try to help a stab victim. If you touch the person your DNA will be on them and it's possible that you could be implicated for the murder.
" You can't plant a fingerprint."
Actually you can. Anyone can make a mold of your finger and then use that to plant your prints around. This was just in a case on Dateline where a guy got a fingerprint from someone else out of plumbers putty in their house and made a wax mold to plant a bloody print. The only way the police knew the print was planted was because of how the blood was on the opposite part of the ridges that it would be on in a real fingerprint.
I'm certanly no expert, but I understand it's extreamly more difficult to prove guilt based on DNA evidence. It's more often used to prove innocence or provide that shadow of a doubt. This technology could greatly help in lowering the chance of someone being falsely prosocuted for a crime.
Teach someone to use the net and they won't bother you for weeks; show them Slashdot and you may never see them again.
think how many times you've been finger printed...cashing checks, getting a driver's liscense...
Zero. Where do you live, and what bank are you with?
A check might have your fingerprints on it, but I've never heard of anyone being asked for fingerprints to cash one. And I got my driver's license without giving fingerprints.
Other comments talk about fingerprints being required to get jobs. Apart from military jobs, who would require this?
Can a person claim exclusive copy rights on their own DNA? Google turns up some firms offering such protection to celebrities. Are they just a scam? Could gene sequencers be classified as circumvention devices under DMCA?
It's called CODIS (COmbined DNA Index System) and it's used to cross reference DNA in unsolved cases throughout the country. Criminal DNA samples are taken and entered into the system. The biggest reason that the system doesn't get more matches (and it does get matches - the press conferences in these cases just don't mention CODIS) is that most jurisdictions are still using different tests (that could have changed by now) and that the backlog of samples to process and enter into the system is so huge that it would take years just to catch up on the California data only, even if nothing new was added.
My $0.02 is that this is a great system to have around, but I'm sure that everyone else will have their own opinions on that.
Btw, since this uses trace amounts of DNA (I'm not sure where it comes from - only the bottom layers of skin cells are nucleated...glands, maybe?) why do they say there is less risk of contamination? Wouldn't it be greater? What if you just shook hands with someone - especially if they had a cut, or hyperhidrosis, or, ugh, had traces of semen or feces (including shed intestinal mucosa) on their hands?
Call me crazy, but raise your hand if you saw Gattaca and thought "Gee, that's a bright and well adjusted future. What do you mean Dystopian?" I sure as hell did.
~
If you need me, I'll be hanging my computer from the
did not give away information about the suspect (eg. race or eye colour
Why? Wouldn't it be cool if they could just take a DNA sample and produce a picture of the person?
Do you care about the security of your wireless mouse?
Large databases are being built up of fingerprint data and now DNA data. The acuracy of this data is at best questionable. Fingerprints are measured at 16 points. From this you do not get 16^16.
I think that if you measured fingerprints to an infinite acuracy you may find the theoretical infinate number of fingerprints required to sustain the myth that no 2 fingerprints are the same but here in the real world we measure a finite number of points and therefore have a finite number of prints and as the database reaches that number there must be mistakes.
The mistakes are already happening with DNA and because this evidence is perported assumed to be infallible innocent people are being arrested.
If this evidence was only used to support other evidence I would see it as a good thing but when it is used as the only evidence then it is very bad.
I think that in the future this DNA witch hunt will be seen for what it is but for now innocent people will be caught up with the guilty.
I love stacking my barbecues in the shed at the end of summer - you can't beat a bit of grill on grill action.
DNA extracted from the region of a fingerprint does not prove that the DNA came from the fingerprint.
Residual DNA coats every surface and depending on the environmental conditions, whether it is inside or outside exposed to the sun, many other sequences will be present.
Current sampling and extraction techniques can not avoid this contamination and if your favorite hangout turns out to be a murder scene, well you are in trouble. While control samples taken at the scene in areas where 'no fingerprints' occur can be taken to test background DNA, it certainly is not foolproof.
Additionally, races and skin types slough skin at different rates and have significant oil-content differences, so there will also likely be a discrepancy in who gets caught. tough luck.
... we have a situation where only 0.000000001% of the people involved actually understand the Science - the rest just assumes "oh - they are probably right" - but if they are not, or have a hidden agenda.... don't let them tell you that one man can't make a difference...
If Google really cared they would fix Android Chrome to reflow text, instead of discriminating
Now, just because one has a copy of someone's DNA, that isn't enough. One diploid copy of human DNA is about 6.6 picograms. If that copu of the genome has been fragmented in one of the regions being amplified, the reaction won't work. True, you could get down into the 50 to 100picogram range for input DNA, but what you're doing is taking the statistics and throwing them out the window. Wheras the kits themselves give odds of matching a random person in the 1 in hundreds of millions to 1 in billions, if you're looking at say 100 copies of degraded DNA (0.6ng, or 600pg), you may only have on average 5 or so copies of intact DNA from the given amplification targets. Now the odds that you only see one allele (say from mom'a side) goes WAY up, because random luck might have caused only 1 or no copies of the other allele (from dad) to survive. The result is, you get an amplification that looks like the person has all one sized fragments for that region, whereas they may really have 2.
Don't get me wrong, I think this technology is probably tremendously useful, and can offer the ability to type people for all sorts of things, such as forensics. I merely wanted to point out some of the potential limitations of such a technique if the sample hasn't been stored well. I have a hard time believing DNA is super stable in black ink...
Just my random thoughts...
I have a hard time believing that they can extract 10 ng of DNA from a fingerprint. A diploid human cell as 6x10e9 bp of DNA. One bp is 660 daltons. Calculating backward, 6x10e9 bp works out to being 6.6 pg of DNA.
So for them to extract 5-10 ng of DNA from a fingerprint, a fingerprint needs to contain between 1000 - 2000 cells. I work with epithelial cells, and a 1000 - 2000 cells is a fairly large patch of cells.
So either they mean that they get 10 ng of PCR amplified DNA (which is possible), but then is hardly representative of the entire genome, or they are using fingerprints from people who are really shedding skin!
I think you've taken a lot of urban legend and stuffed it into one big slashdot post.
DNA analysis by RFLP (Restriction Fragment Length Polymorphism) is very, very accurate. This is how it works:
On your DNA, you've got lots of little molecules. These molecules form a sequence. Every so often, there will be certain repeated sequences by chance.
Restriction Enzymes locate these sequences and go *snip*! They break down the DNA at these specific sequence points.
The DNA is then run through a gel - the smaller fragments go farther through the gel. The gel is then analyzed for the particular pattern of fragments in the gel.
In case you didn't catch the variabilty associated with all of this - these restriction fragments snip only at repeated sequences, repeated sequences which occur at random in our DNA. The chances of two people having exactly the same combination of restriction-snipped fragments is so so so so *so* small it is difficult to express in numbers - think about what you're saying.
There ARE cases of fallible DNA tests - DNA tests that aren't done properly, etc. But few people are ever jailed wrongly because of properly collected DNA evidence.
My biggest issue with DNA evidence is that it only proves that the suspect was at the scene, not that he commited the crime.
What if I touch a gun and 5 min later you kill someone with it. They might find my DNA but your fingerprints. If they are really good they will find both sources. Seems like resonable doubt to me. If they just go by finger print, closed case. Guess what I am saying is that they might find a bunch of DNA and have to release guilty people.
You say:
This is only true if you get a sufficiently large number of fragments. If you're analyzing someone's entire genome, of course you're right -- the only possible way to get an identical "DNA fingerprint" is on identical twins. But in fact the number of fragments analyzed is fairly small, in the thousands; which means it's possible to get the same analysis out of several million unrelated people, and a much smaller number of closely related people. Considering how many crimes are committed by one family member against another, this is a real concern.
I'm all for DNA analysis as a forensic tool, since it's currently the most accurate tool we have for placing a suspect at the scene of a crime. But it's a long way from perfect. Presumably, as the technology improves and it becomes practical to analyze larger sequences faster, it will get better.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.