Current Focus

Algorithms for Single-particle Structure Determination

X-ray Free Electron Lasers (XFEL's) promise to revolutionize structure determination as lasers transformed spectroscopy.  By delivering 200-fs pulses containing 1012 coherent photons, XFEL's will provide 11 orders of magnitude higher brightness than currently available.    Photons are anticipated from the LCLS, world's first XFEL located at the Stanford Linear Collider in less than a year, with subsequent instruments due to begin operation in Japan (2011), Germany (2014), and Switzerland (2016), respectively.

One of the most exciting applications of such instruments is the study of the structure and dynamics of single particles, ranging from aerosols and nanoparticles to individual biological molecules.  We are developing the key algorithms needed to study the structure of individual macromolecules, viruses, cells, and nanoparticles in four dimensions by single-particle X-ray scattering.

So-called "Diffract-and-Destroy" approaches successively expose a train of particles to intense, ultra-short X-ray pulses, and collect diffraction "snapshots" from single particles of unknown orientation before the object is destroyed by the pulse (see figure).  The diffraction patterns must be oriented relative to each other and used to reconstruct the 3-D diffraction volume in reciprocal space.  The object structure can then be determined via iterative "phasing algorithms."  Interrogating single particles would obviate the need for crystals, and a short data collection window would "out-run" radiation damage.  This would represent a transformative advance in structure determination, with potential impact ranging from nanoparticles and cells to individual macromolecules.



  Department of Physics

The alluring attributes of "diffract and destroy" single-particle approaches are matched by significant experimental and algorithmic challenges.  Substantial progress has been made on the experimental side.  On the algorithmic side, we were the first group to demonstrate the key algorithms needed for reconstructing the molecular structure from a random collection of signal-starved diffraction "snapshots" of a single biological molecule. (See: Structure by Fleeting Illumination of Faint Spinning Objects in Flight.)

More generally, we are developing powerful manifold-embedding techniques to map the structure and conformations of a class of heterogeneous objects from random snapshots of members of the class .  This hitherto unanticipated possibility should allow the study of molecular reactions, chromosome, and perhaps even whole cells in four dimensions.  Example applications include mapping the conformational continuum of biomolecules in reaction, the study of "living" cells, the elucidation of the effect of particle shape in catalysis, and ultra-low-dose tomography of beating hearts and breathing lungs.

Single-particle structure determination by short-pulse X-ray scattering

(Graphic from Gaffney & Chapman with permission.)








Research Projects

Course Outline

Current Focus

Job Opportunities

Recent Highlights