Conquering 3D Single Particle X-ray Diffraction Imaging
- Reference number
- ICA12-0064
- Start and end dates
- 130901-170831
- Amount granted
- 3 000 000 SEK
- Administrative organization
- Uppsala University
- Research area
- Life Sciences
Summary
The advent of X-ray free-electron lasers has provided users with beams of unprecedented peak brightness, a billion times brighter than previous sources, and briefness with pulse duration on only a few femtoseconds. These properties made the concept of diffraction before destruction possible. In such an experiment a sample is subject to such an intense X-ray beam that it explodes, but thanks to the extremely short pulse, which is faster than the explosion, an image of the intact sample can be recovered from the scattered photons, much like flash photography can freeze a subject. Single particle diffraction at X-ray free-electron lasers using the diffraction before destruction technique is booming and there have been several promising results published. I was part of the team that imaged the first virus using the technique, published last year in Nature. The next objective is obtaining three dimensional images of reproducible biological systems, such as virus particles, by combining the data from multiple diffraction images, each obtained from a similar sample. There are several promising algorithms proposed to achieve this, but they need to be further developed and refined with experimental data. Making three dimensional single particle image a routine technique is the goal of this project. This will provide an invaluable new imaging tool with broad impact in several areas of biology particularly celular, molecular and structural biology, complementing existing techniques.
Popular science description
X-ray lasers and a new kind of X-ray source which is a billion times brighter than existing sources. It is so powerful that when focused on a micron spot it has the same power density than all the sunlight hitting the same concentrated in a square millimeter. Such new tools enable new experiments. One of them is imaging samples at the nanoscale level by using the fact that objects when hit by X-rays will scatter some of them and those scattered X-rays can be used to recover the structure of the object. If one has a reproducible sample is then possible to obtain a three-dimensional picture of the object by obtaining many two dimensional pictures of it from different directions and combining them in a single three-dimensional one. This was one of the dreams that led to the construction of these advanced machines and this project aims to make that dream a reality. The main obstacle which currently prevents the realisation of this dream is the lack of sufficiently advanced data analysis capabilities to combine the images that are gathered during the experiments in a single cohesive three-dimensional picture. This project will try all proposed algorithms to achieve this and test them with real experimental data, in collaboration with experimentalists. The best performing algorithms will then be picked and tweaked up to their maximum performance. New ideas from a new branch in applied mathematics, compressed sensing, will then be used to try to enhance the best algorithms and push them beyond their limits. Finally the whole software infrastructure will be connected and tested in an experimental facility which should allow its use by non-experts in diffraction imaging, such as researchers in other fields and industry. This new tool will make it possible to obtain 3D pictures with nanometer resolution of a large variety of samples, making with extremely useful in a wide variety of domains, from academic to industry, and is expected to become an important technique complementing existing well established imaging techniques such a electron microscopy, and atomic force microscopy.