The fine structural details of individual particles can be difficult to discern from raw images taken on an electron microscope. This is due to the low signal-to-noise ratio typical for imaging of radiation-sensitive biological samples. It becomes even more challenging when the particles are conformationally or compositionally heterogeneous.
2D class averaging enhances signal, increasing information that can be gleaned.
In this process, individual particles are selected from TEM images, aligned relative to each other, and then computationally classified based on apparent similarities. In this way each particle class becomes an average of many similar particles, which enhances the overall signal. These 2D class averages provide a rich source of information on basic particle features, domain structures, and can also help in understanding the degree of particle flexibility.