Taking snapshots of biomolecules such as proteins and DNA is fraught with difficulty. The approach made famous by Bragg, Franklin, Watson and Crick and others is to create a crystal of the molecule in question and bombard it with x-rays or high energy electrons. The subsequent diffraction pattern can then be used to determine the shape of the molecules in the crystal.

The trouble is that these kinds of massive, complex molecules can take all kinds of different shapes and the diffraction process gives only an average of all the molecules present. And that's only if the molecules happen to crystallise. The trouble for molecular biologists is that many biomolecules, membrane proteins in particular, won't form into crystals.

So physicists have been working on various techniques for photographing individual molecules. The trouble is that the x-ray or high energy electrons used to make the pictures end up destroying the molecules in question. And since it can take some 10,000 images to build up an image that is clear enough to show structural features, each of these images must be of a different molecule. So once again, the various structural differences between molecules get smeared out.

Now Matthias Germann and buddies at the University of Zurich have a different approach. Instead of high energy electrons, they've created holograms of DNA strands using a coherent beam of low energy electrons (although why this approach hasn't proved fruitful in the past isn't clear).

Their results show that at certain energies, DNA strands are remarkably robust to low energy electrons. "DNA withstands irradiation by coherent low energy electrons and remains unperturbed even after a total dose of at least 5 orders of magnitude larger than the permissible dose in X-ray or high energy electron imaging," say the team.

Yep, that's five orders of magnitude more.

What this suggests is that if you choose electron beams of just right energy--Germann and co say 60eV does the trick--then it becomes possible to take decent snapshots of DNA molecules without destroying them.

That could suddenly make it possible to determine the structure of numerous proteins and other biomolecules that have so far avoided the scrutiny of molecular biologists. Not to mention cheap sequencing simple by pulling a strand of DNA past the right kind of video camera.

Ref: arxiv.org/abs/0910.1499: Non-destructive Imaging of Individual Bio-Molecules