DATA HETEROGENEITY IN SINGLE PARTICLE IMAGING EXPERIMENT WITH X-RAY FREE ELECTRON LASER

Abstract

The X-ray free-electron laser (XFEL) is an emerging technology that may overcome the damage-limited resolution in single particle imaging (SPI). The difficulty in SPI data analysis mainly comes from two kinds of latent unmeasured variables in datasets: orientational heterogeneity caused by randomly oriented particles and structural heterogeneity within an ensemble.

While the missing-orientation problem can be solved by the expand-maximize-compress(EMC) algorithm, the addition of structural heterogeneity quickly complicates structural recovery. In this thesis, the EMC algorithm is extended to reconstruct multiple diffraction intensities simultaneously. This mmEMC algorithm could also be used to single model case for improving signal by purifying the dataset in silico. We also propose an important, but sorely absent, information-theoretic approach to validating SPI reconstruction. Using related ideas, we probe the structural differences in a large ensemble and probabilistically describe the whole structural ensemble.