Electron microscopy (EM) offers a means by which to study large protein complexes in a near-native environment without the need for crystallization. The protein quantity necessary to arrive at a structure is additionally significantly less than the amounts needed by crystallographers to screen crystallization conditions, and 1000-fold less than what is required for nuclear magnetic resonance (NMR).

The largest drawback to this technique is that the high resolutions achieved by NMR and crystallography (better than 5 Angstroms) can only be achieved using EM with exceptionally well-ordered and stable molecules or molecules that have a high degree of symmetry. Such resolutions can also be achieved by EM in conjunction with 2D crystallography, although the sample preparation for this type of study requires ordering of a given sample into a two dimensional lattice, bringing about some of the same difficulties experienced by X-ray crystallographers.

The ability to observe molecular complexes in conditions similar to those found within the cell, however, is the most important aspect of EM. Because macromolecules do not have to be crystallized for EM analysis, a protein is free to adopt any of the conformations it may assume during dynamic processes within the cell. These different conformational states can be segregated into groups, providing a more complete description of the molecular movements necessary for function. Such analysis of heterogeneity is also applicable in examination of varying assembly or maturation states, and additionally removes the requirement of extreme sample purity in analysis, which is crucial for both crystallography and NMR. The low-resolution structures of large or dynamic molecular complexes solved by EM can be combined with high-resolution structures of substituent components or fragments solved by either crystallography or NMR. This marriage of techniques enables structural biologists to explore the structural architecture and conformational mobility that underlie many biological processes.