DNA and Lasers

[foodchain]

Now giving how mass spectrometry works I would like to purpose a question that popped in my head yesterday. Say you have a certain chunk of DNA, would it be possible to use say some type of chemical to elongate it in a tube, and then basically run the length of it with say a type of laser to figure out what each base is?

 

Basically you would put an amount of DNA in a tube or vial type device for example, and then say some type of farmed or culture enzyme would basically uncoil it and make it straight, maybe some type of chemical strip in the vial possibly, and then after the strand is fully elongated you could run a type of laser along it to figure out what each base is?

[Klaynos]

A laser wouldn't really work as it is just a single wavelength. You could use a white light source, but how big is each "base"? As the resolution is wavelength limited, as well has having a limited spot size...

[John Cuthber]

You can olny focus the laser down to a spot size of roughly half a micron.This is huge compared to the individual base pair (about a third of a nanometre). If you had an Xray laser then you could possibly do what you suggest but sequencing DNA isn't that difficult; it's a lot easier than getting an Xray laser. Even if you could get the xray laser you would still have problems. First Xrays destroy DNA. Secondly the signal from just one DNA molecule would be too small to see properly. You would need lots of molecules all lined up and I can't see that being easy. Even crystaline DNA isn't that well aligned.

[Reaper]

Yeah, pretty much you really can't use a laser to do anything with it. Anything beyond ultraviolet is way too energetic, and anything below that is way too big for the DNA molecule.

 

We usually use electron microscopes to see details of DNA strands.

[foodchain]

Well thanks, I had no idea really. I got thinking on it simply from say mutation rates of bacteria or more or less thinking of some way to possibly very rapidly discover bases on a certain length of DNA, I guess the laser is out of the question though.

[pioneer]

Actually an easier way uses NMR or nuclear magnetic resonance. There are only two base pairs that have either 2 or 3 hydrogen bonds. If you know how many H-bonds you know which base pair. You scan to find the 2233233232323... sequence. The rest of the DNA and the bases are implied in the sequence. This tells us the base pair but is still ambiguous as to which base is on which of the two stands. There is another tell, that can also use the NMR, and would need to immediately follow the first scan. It has to do with only one base, of each base pair, having extra nonbonding H. The first scan looks for 223322 and the second 11212121

[John Cuthber]

Well, I may have missed a few things but here's my thoughts.

Last time I looked NMR used radio waves to do the spectroscopy. I guess they might be into the microwave region by now so that's a tacit spatial resolution of a centimetre or so. With some clever tricks you can get MRI images with resolutions of a milimetre or so.

I thought the half micron that you could get with a laser was hopeless. Half a milimetre is a whole lot worse.

 

It gets even worse still if you think about the signal to noise ratio. In general terms the higher the energy photons you use, the easier it is to see them against the background noise.

It's relatively easy to see the effect of a single gamma ray photon form a decaying atom

It's possible, but very difficult to observe visible fluorescence emission from a single atom or molecule.

I doubt that anyone has even tried to do this experiment with an infrared transition.

Good luck to anyone trying to do it with microwaves.

I suspect there are also problems with relaxation times.

 

Or we are back to the problem of trying to line up lots and lots of DNA in perfect register.

 

Guess why they chose the techniques they do use for DNA sequencing.

[Klaynos]

NMR is not a spectroscopy method, and never has been.

 

MRI and NMR are the same thing.

 

You nearly always (if not always) get pairs of photos from decaying atoms. Which I can tell you are non-trivial to find, and show they are coming from your sample sat in the middle of the desk :| Alot of the time noise etc depends on the detector more than the energy of the photon you are detecting, and the amount of other stuff emitting in that region (why IR and visible is so annoying). I've had to angle experiments in very strange ways because my detector pair tuned to a specific energy has been going crazy when going anywhere near one region of a room.