IDeAS Seminar: Reconstructing 3D protein crystal intensity from unoriented sparse diffraction patterns

Ti-Yen Lan, Cornell University
Dec 6 2017 - 2:30pm
Event type: 
224 Fine Hall

The femtoseconds long pulses of an X-ray free electron laser (XFEL) enable the measurement to outrun the irreversible radiation damage. This concept of `diffract before destroy’ inspires new methods such as serial crystallography, which determines 3D protein structures by merging 2D snapshots of microcrystals collected at random orientations. In recent years, there has been a growing interest to apply this technique to the synchrotron storage ring sources because of their wider availability. For very small crystals, however, radiation damage occurs before sufficient numbers of photons are diffracted to determine the orientation of the crystal. The challenge is to merge data from a large number of such “sparse” frames in order to measure the full reciprocal space intensity.

In this talk I will discuss our effort to develop an analysis method to analyze the unoriented sparse crystal diffraction patterns. Using the EMC algorithm (Loh & Elser, 2009), we reconstruct the 3D crystal intensity by iteratively maximizing the likelihood function of the crystal orientations. I will demonstrate the ability of our method to analyze sparse patterns through a couple of proof-of-concept experiments with increasing complexity. Finally I will present our recent progress in applying the EMC algorithm to a serial crystallography dataset taken at a storage ring source. 

Ti-Yen Lan received his B.S. degree in Physics from National Taiwan University, Taiwan, in 2012. He is now a Ph.D. candidate in Theoretical Physics working with Professor Veit Elser at Cornell University. He is interested in developing methods to extract information from noisy image data through the modeling of the image formation process.