Advanced image processing software for membrane protein structure determination

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): We will develop a software system to improve the speed, efficiency and resolution of membrane protein structure determination. While membrane proteins comprise the majority of pharmaceutical drug targets, lack of structural information for human membrane proteins poses a roadblock for structure-based drug design. Conventional protein structure determination by X-ray or NMR methods faces serious challenges when applied to membrane proteins. Electron crystallography instead uses high-resolution electron microscopes to image two-dimensional crystals of lipid membrane-reconstituted membrane proteins. Unfortunately, even with electron crystallography the structure resolution required for drug design has met with difficulties in sample preparation and image processing. We propose to develop a new image processing system that will modernize the structure reconstruction from electron crystallography images by including several recently developed mathematical algorithms, thereby improving the resolution of the reconstructions. Our system will automate the processing, which will enable competitive processing times. And, most importantly, our new algorithms will greatly improve the efficiency of the method, by being able to process images that previously could not be used. Phase I work will prove feasibility by implementing three software extensions at command line level and demonstrating improved resolution. Phase II will further refine the mathematic modules and implement them is a user friendly GUI. This software will be the technological basis for EON Structures to enable high-throughput, high-resolution membrane protein structure determination for pharmaceutical research and structure-based drug design. Membrane proteins represent the main target for drugs currently under pharmaceutical development. Structural based drug design is one of the most powerful methods of drug development, but structures for membrane proteins have been extremely difficult to produce. We will test new methods for computer image processing which should allow us to greatly advance the speed and resolution of determining the 3D structure of membrane proteins. This project will have significant commercial impact both through direct sale and license of the new software, and by enabling our use of these tools for contract based membrane protein structure determination for pharmaceutical companies.