Towards predictive structure-based models of evolved drug resistance

The evolution of drug resistance in malaria continues to be a widespread concern. Many of these drugs target key proteins such as dihydrofolate reductase (DHFR). However in malaria, the structural plasticity of DHFR allows it to maintain its active site and catalytic activity, while resisting drug binding. One way to better understand this process is through the appreciation of DHFR structural evolution in general, and then use in silico evolution to model both the drug docking and the likely amino acid changes in DHFR that will occur as a result. Using a comprehensive phylogenetic analysis of DHFR, we have generated a method that generates variant DHFR proteins and sequentially scores the docking of the natural cofactor NADPH and the known anti-malarial drug pyrimethamine in order to determine fitness. Iteration of this process allows the opportunity to model the coevolutionary processes involved with drug resistance and to predict responses to pharmaceuticals in advance of their use in the field.