Automated Orienting of Water Molecules in Neutron Crystal Structures

Sammanfattning: A protein’s function is directly related to its structure, which is in a water medium where it is affected by hydrogen bonds and the hydrophobic effect. These interactions are in turn dependent on the water molecules’ orientations around the protein. Therefore, it is vital to have correct orientations for the water molecules. Such information can be obtained by neutron crystallography. However, even in such structures, the correct orientation of water requires a manual evaluation and possible re-orientation of each water molecule. This is a tedious and time-consuming procedure since proteins typically contain hundreds of water molecules in their lattice and can have several sub-units. Therefore, we have here tried to develop a method that reliably automatizes the orienting of the water molecules in a simple and relatively fast way. As test cases, we used the proteins galectin-3C, rubredoxin and pyrophosphatase. We evaluated the water molecules' orientations both quantitatively, with RSCC values, and qualitatively, by studying the orientations in density maps. We have optimized the refinement by varying the optimization methods and refinement parameters, thus finding the settings that yielded the best results in terms of time and performance. In particular, we have constructed two scripts that identify and re-orient inadequately oriented water molecules. Ultimately, we performed the refinement and re-orienting using only neutron data. We show that our approach yields improved orientations of the water molecules for all three proteins, in a shorter time than a manual orientation.