When a car hits something, it often leaves some paint behind. I have come out to find mysterious paint scrapes on my car once or twice, with no note or other explanation of what happened. More serious are hit and run collisions, where the paint left behind is sometimes used to narrow down the make and model of the responsible car.

DNA fingerprinting is now commonly used as physical evidence to apprehend criminals, and some areas collect DNA on everyone arrested, building up a library for future use. Why not put DNA in car paint? It is very tough and persists for years in the environment. It would be very simple to put together an inidividual identifier for each car - one would use a mix of one or more DNA "letters", each letter being a 50 nucleotide stretch. More letters could be introduced as needed. The letters would be premade and an individual mix sprayed on the final coat of paint. Records would link the DNA tag to the car serial number. Paint left on the scene would undergo DNA fingerprinting just as blood and hair does now.
-- bungston, Jun 26 2003

Chips in License Plates http://www.halfbake..._20license_20plates
This idea got me thinking about hit and run accidents. [bungston, Oct 05 2004, last modified Oct 17 2004]

Spray-on DataDot identification http://www.datadotd...gy_ourtechs_dot.htm
Making this compulsory for new cars has been toyed with locally [Adze, Sep 17 2005]

DNA is reasonable stable, but on the surface of a car exposed to high temperatures, salt spray, UV light etc. it won't last very long. Anyway you wouldn't be allowed to sell a car like that in Europe. They don't like genetically modified cars there.

The manufacturers could still put some additive into the paint (marked fibers like in counterfeit proof paper?). The additive could encode manufacturer, model, year to cut down the number of suspects.
-- kbecker, Jun 26 2003


Current paint analysis lets investigators hone in on make and year of car just with a few scrapes.

Methinks taking a production line and mixing a custom, DNA-encoded paint for each car would add a lot of expense to each car. Would you pay, maybe $1000.00 more for a car with this feature?
-- Cedar Park, Jun 26 2003


[CP] DNA wouldn't work (see earlier note), but little flitters of holographic metal foil would be cheap. Not cheap enough to tag each car individually because of the way the painting process is set up, but cheap enough for model and year. It would cost pennies per car and make the paint analysis much cheaper.
-- kbecker, Jun 26 2003


I heard something recently about a way to make small RFID tags that were sprayed onto the underside of a car to identify it.

I think your best bet with this plan would be to use some kind of chemical marker or the holographic foil like kbecker said. You could have a few letters that identify the make and model like //VW BTL// or print part of the VIN.
-- discontinuuity, Sep 16 2005


It occurred to me that this DNA method could be used to track batches of prescription medicine - for example, when people overdose on oxycontin, where did that oxycontin come from? DNA sequence could also be used to certify authenticity of branded medicines. DNA is nontoxic and the sequence would not be hard to alter from batch to batch.
-- bungston, Nov 30 2009


// DNA is nontoxic// Not true. Many people get congenital disorders from DNA. I recommend not going anywhere near the stuff.
-- swimswim, Nov 30 2009


This would work. DNA is indeed fairly stable, and if it were incorporated into the paint (and if the paint were chosen appropriately), it would probably survive well beyond the life of the car. The main enemies of DNA (other than enzymatic degradation) are UV, water and metal ions; it would be protected from these inside the paint. It could be released by dissolving the paint in a suitable solvent.

It would also be pretty cheap. Let's assume that, from a scrape amounting to 1 square centimetre of paint, you would want to recover one picogram of DNA. (This is bucketloads, of a small DNA fragment.) So, if the car has 10 square metres of paint, this is 10^5 picograms, or 0.1 micrograms of DNA.

At the normal DNA synthesis scales, you'd be looking at a few tens of pounds for that amount of DNA. However, given that you'd be making very large amounts of each DNA molecule, the cost per car would probably be under a pound, for the DNA.

It could be added to the paint close to the spray nozzle (so each car only gets one code). Alternatively, it could be applied on top of the primer/undercoat in a simple aqueous spray, then oversprayed with regular paint.

So, doable. (+)
-- MaxwellBuchanan, Nov 30 2009



random, halfbakery