Aggregation is both a blessing and a curse to the biochemist seeking to purify and use recombinant proteins. Reversible aggregation/precipitation induced by agents such as polyethylene glycol and salts can be used to purify proteins. However, proteins often undergo uncontrolled and irreversible aggregation that prevents regeneration of activity. This article summarizes methods to measure aggregation and to determine the intrinsic solubility of proteins. Much of this research has been driven by the need to formulate medically useful proteins so that they remain soluble, and to develop treatments for diseases that involve protein aggregation, such as Alzheimer's and other neurodegenerative diseases.
Theoretical treatments of the effects of co‐solvents on the solubility of proteins, including using empirical data to derive
atomic solvation parameters
(ASP), may be useful for some applications. Mathematical models and protein phase diagrams have been developed for common precipitants, although there is no common basis for their
a priori
use with a given protein. There are also some general rules for designing mutants with reduced aggregation tendencies. To this end, computational methods may be used to estimate the solvent exposure of residues in 3D‐structures, predict surface patches that may dictate protein–protein interactions, and to model the 3D structure.