The first crystal structure of a GABAA receptor has appeared (Miller & Aricescu 2014) and further subtypes and ligand bound structures will follow soon. This will have a big impact on our work as we no longer have to rely on homology models for structural information. The emphasis of our work will shift from predicting gross differences between distant homologues and GABAA receptors, to predicting subtle differences between subtypes. We hope to develop bound state structures of subtypes with their selective ligands, such as the α6β3γ2 selective 'compound 6' from Varagic et al. Br. J. Pharmacology 2013.