Molecular Replacement

Molecular Replacement, Jason Schmidberger and A. Prof. Ashley Buckle, Department of Biochemistry and Molecular Biology.
One of the major challenges for Protein Crystallographers is to solve the ‘Phase Problem’ for the diffraction data that they collect from their protein crystals, thereby solving their three dimensional structures. A very useful and common method of doing this is Molecular Replacement (MR), where a homologous protein’s structure is used as a ‘starting point’ in the structure determination process. Unfortunately this can be a difficult process with model (homolog) bias being a big issue. Small variations parts of homolog and target protein structures can significantly influence the success or failure of a MR calculation. For this reason it is very useful to try as many different homolog structures as possible, while also varying a number of other parameters essential to the MR calculation.

This is where the Monash Green SPONGE Condor pool has aided Crystallographers' research greatly. Through it, researchers have been given access to over 1000 separate computer cores and because their MR calculations are ‘embarrassingly parallel’, they are able to construct incredibly large parameter sweeps, optimising the chance of getting a favourable result. One area that they have developed in particular, is a process that links together Normal Model Analysis (NMA – a computational method to identify possible flexibility in protein structures), with a 3D structural alignment process coupled with a sieving (or trimming) process. This pipeline will take a single protein chain and can produce 10’s to 100’s of variations on that structure (Figures below), which can all be paired with further parameter sweeps and all run on the Monash SPONGE Grid.