My work focuses on making the best use of X-radiation as a tool to determine macromolecular structure. Software plays an important role in collecting the best possible data and extracting the most accurate information. Beginning in 1999 at the Stanford Synchrotron Radiation Laboratory (SSRL), I was part of the development teams for BLU-ICEand Web-Ice, now-popular software systems that manage experiments within a single interface. Moving to my present position in the Computational Crystallography Initiative at Lawrence Berkeley National Laboratory in 2000, I helped develop CCTBX, a powerful open-source software toolbox forming the basis for the structure refinement program PHENIX. Collaboration with protein crystallography beamline groups here at the Advanced Light Source led to the development of LABELIT to handle repetitive image analysis without needing human guidance.
Current challenges center on ultrafast detectors that are revolutionizing experiments, producing massive datasets that require profoundly new computational approaches. We are targeting new algorithms for physics-based modeling of the diffraction data, and are experimenting with new GPU hardware. In collaboration with international partners, we are developing new NIH-sponsored packages to process synchrotron data (DIALS) and free-electron laser data (cctbx.xfel). XFEL experiments are also contributing to an LDRD-funded effort to understand the photosynthetic mechanism of water splitting.