Techniques for Obtaining High Resolution Information about Periodic Arrays of Biological Macromolecules in the Electron Microscope
In the past six years considerable effort has been expended in EM structure analysis of periodic arrays of biological macromolecules. Although sparked by the success of the moderately high resolution structure analysis of purple membrane J most of this work has been limited in resolution to anywhere from 30 to 15 A. Although there is much to learn out to this range of resolution, especially in three dimensions, there is a wealth of structural information awaiting us beyond it. Our ultimate goal must, of course, be information in the range of 3 A, which for proteins is enough to allow us to trace the path of the polypeptide chain. In fact, much of this information is accessible by electron microscopy, providing that the periodic arrays are ordered to this extent. However, great care must be taken to optimize all aspects of microscopy and image processing in order for structural information of a resolution close to the instrumental limitation to be obtained. Following are some of the more important factors which will determine success or failure of an attempt at high resolution imaging, in this case assumed to be of an untilted specimen. While some of these factors have been known for years by workers in the field of high resolution electron microscopy, others are problems unique to imaging biological macromolecular arrays. The recent solution of some of these problems involves the use of probability distributions to describe image phases.