Princeton University   |   Department of Molecular Biology

X-ray diffraction image
v-FLIP MC159 X-ray diffraction from data set at
X29 beam line at NSLS-I
X-ray diffraction image
Diffraction image for tetragonal lysozyme using
the RAXIS-IV++ detector and RuH3R home source

Once you've obtained crystals of a suitable size that diffract fairly well, the collection of X-ray data is a straightforward parts of the process. The facility has a new Rigaku MicroMax-007HF microfocus rotating anode generator operating at 1.2 kW and generating CuKα x-rays (λ = 1.54 Å) which are the best choice for macromolecular samples. Varimax HF multilayer optics focus and collimate the X-ray beam and a Dectris Pilatus3 R 300K hybrid pixel array detector records the diffraction data. The new system - installed June 2017 - is effectively 50x brighter than the outgoing system. We can now do in minutes what used to take hours, and we can also push the limits of weaker diffractors. The current in-house source is viable for data collection for moderately-diffracting for crystals that have been cryo-cooled to 100 Kelvin using our Oxford Cryostream model 700. Protein crystals accumulate radiation damage quite quickly at room temperature but this is radically reduced by flash-cooling them to 100 K in a nitrogen gas stream. Data collection on the in-house source takes between one hour to one day.

For an example of what data collection on a well-diffracting crystal looks like, here's a video of a 90° data collection (100 sec/0.5°) crammed into 7 seconds from our older in-house source. Proteinase K is a strong diffractor and this data extends well past the edge of the detector at 1.6 Å resolution. You could collect that sort of data in 15 minutes on our current system.

For straightforward cases (reasonable diffractors, available homologous structures) we can determine structures using data collected in-house. For structures with enough sulfurs (Met, Cys) and strongly-diffracting crystals we can now attempt S-SAD experimental phasing in-house, or we can use other anomalous scatterers (Iodide, Ho3+ substituting for Ca2+, etc). For other cases (novel structures and/or weak diffractors) we must use the ultra-bright tunable X-ray sources at synchrotrons. In the last couple of years we've used the Cornell High Energy Synchrotron Source (CHESS) beam lines A1 and F1, and the AMX beamline of NSLS-II at Brookhaven National Lab.

For the in-house source we support the HKL3000 package for data collection and processing.

HKL3000 screen shot
HKL3000. Click image for larger version

Since any individual data collection package has its strengths and weaknesses we also support XDS, MOSFLM, autoProc and Xia2 processing programs for data collected either in-house or at synchrotrons.

We can also collect small-molecule data sets to a maximum resolution of 0.83 Å on the new system - complete structures are usually possible in a couple of hours and initial structures possible within 10 minutes using the CrysAlisPro software.