As if focusing high brilliance, high energy photons from a synchrotron source onto a temperature-controlled solution of biological material weren't difficult enough, the data from the experiment, at first glance, don't mean a whole heck of a lot. That's where Dr. Mardis comes in.
In X-ray crystallography, the diffraction pattern can be converted into a single structure of a molecule. However in solution, a molecule is moving, stretching, vibrating, and spinning; and each orientation contributes differently to the 1D pattern of the X-Ray scattering experiment. With molecular dynamics, Dr. Mardis determines a whole bunch (say, 1000) of structures that a molecule can exist in and then predicts the X-ray scattering pattern for each structure. After averaging all those patterns together, she gets a final graph that looks pretty darn close to what her experimental colleagues find in the lab.
Dr. Mardis recently spoke about this publication at a departmental seminar. Spending the summer at a national lab (and the hope of playing video games at break-neck speeds) is just one of the advantages of working in her research group. If you'd like to read the article, you can find it here if you have access to the journal or contact Dr. Mardis for a copy (this link sends you to the CSU Chemistry and Physics faculty contact info).