LipID: A Lipid Annotation Tool for Membrane Proteins (360G-Wellcome-208361_Z_17_Z)

Interactions of membrane proteins with lipids are important in their stability, regulation, and targeting. These interactions are of biomedical importance given the roles of membrane proteins in disease and as drug targets; with drugs often acting via lipid binding sites. We will develop a computational pipeline and database for molecular simulations to predict the structure, specificity, and affinity of membrane protein/lipid interactions. This will provide a unique 'computational biochemistry' resource for functional annotation of membrane protein structures, identifying potentially druggable sites. Building upon our MemProtMD methodology (http://sbcb.bioch.ox.ac.uk/memprotmd/), we will provide a server enabling simulation-based predictions of lipid binding sites to membrane protein structures, defining the structural basis of lipid interactions for each protein alongside predictions of lipid specificity and affinities. We have successfully established proof-of-principle methodologies which allow us to predict and explore free energy landscapes of membrane protein-lipid interactions. This technology needs methodological development to improve the accuracy, applicability, and robustness of predictions of lipid-protein interactions. We will develop a high-throughput software pipeline for lipid annotation of membrane proteins (LipID) for both known and new structures of two major groupings of membrane protein: (i) receptors, channels, and transporters from humans; and (ii) membrane proteins from pathogenic bacteria.