It is well known that the double helix structure of DNA was determined by the x-ray diffraction method.

But it always baffles me. DNA strands are not simple crystals. They are not periodic, not at all.

So how is it possible to reveal the double helix structure by x-ray diffraction?

  • 2
    $\begingroup$ To get a clear diffraction image one needs crystalized DNA (but only relatively short pieces). Randomly oriented samples (powder method) will still give valuable information about symmetries, but they won't give detailed structural information. The problem does, by the way, apply equally to proteins. If a protein doesn't crystalize well, it is not possible to get high quality x-ray diffraction information. And to make this clear, the sequence of DNA can not be read with x-ray diffraction. $\endgroup$ – CuriousOne Jun 6 '16 at 14:04
  • 3
    $\begingroup$ You can crystalize DNA (and many proteins), making a periodic structure. The main challenges are (1) the large number of atoms in the 'basis' to sort out, and (2) crystallizing a large enough sample to get enough signal from (and lots of signal is needed for (1)). $\endgroup$ – Jon Custer Jun 6 '16 at 14:05
  • $\begingroup$ Here is a good, not too technical, description. dnalc.org/view/… $\endgroup$ – Farcher Jun 6 '16 at 14:54
  • $\begingroup$ You may also find this link useful. To get protein structure, 1) refine the protein to desired species, 2) crystalline the protein (Hardest part), 3) get the diffraction data. You now have intensity part of Fourier transform of protein crystal. The protein structure can be retrived by solving the 2D phase problem. This is known as diffracitve imaging. There will be aliasing effects, comes from the diffraction by sample boundaries. Single protein can be imaged by diffracitve imaging however flux requirement is huge. $\endgroup$ – hsinghal Jun 6 '16 at 18:36

Contrary to your claim, DNA strands do have sufficient periodicity to produce an x-ray diffraction pattern: in essence, when seen from the side, the helix collapses to two sets of diagonal lines, each of which acts as a diffraction grating, producing as an end result a cross formed from two diagonal sets of dots.

Moreover, this simple diffraction pattern can be readily produced at home using a laser pointer and a lightbulb filament, as seen in e.g. this question, the YouTube video it links to, and several related videos on YouTube, particularly DNA diffraction with a LASER! || MinuteLaboratory #14.

To scale up from a single molecule to an observable diffraction pattern of the double spiral, the clearest route is to crystallize the DNA, forming a periodic arrangement of DNA strands with the same orientation, so that the diffraction patterns from their helical structures adds up to an observable pattern. If you then make sure that the crystal structure is of identical DNA strands (with the same nucleotide sequences) and that these are synced up, then you can peer into the internal structure of the nucleic bases, but that's not necessary if, like Rosalind Franklin, all you're after is the diffraction pattern from the double helix.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.