Irreversible protein–protein aggregation is associated with several human maladies, including Alzheimer's disease, and poses a challenge to pharmaceutical formulation and preservation efforts. The existence of multiple agglomeration pathways that lead to different final morphologies complicates the ability to determine the aggregation fate of proteins under various reaction conditions. Using lysozyme (3 mg mL⁻¹) as a model protein for dynamic light scattering and circular dichroism studies, we observed dense, non-fibrous aggregation triggered under acidic (pH 4) conditions by a destabilizing “ethanol shock” of 16% (v/v) ethanol. However, ethanol concentrations 15% and below did not lead to aggregation. The aggregation fate of these proteins is thus found to be extremely sensitive to the denaturant concentration. We introduce a stochastic model to explain this dependence based on the denaturant-induced population of a partially unfolded state.

Whether the spatial arrangement of a population influences adaptive evolution has puzzled evolutionary biologists. Theoretical models make contrasting predictions about the probability a beneficial mutation will become fixed in a population for certain topologies like stars, where leaf populations are connected through a hub. To date, these predictions have not been evaluated under realistic conditions. Here, we test the prediction that topology can change the fixation probability both in vitro and in silico by tracking the dynamics of a beneficial mutant under positive selection as it spreads through networks of different topologies. Our results provide empirical support that metapopulation topology can increase the likelihood that a beneficial mutation spreads, broadens the conditions under which this phenomenon is thought to occur, and points the way towards using network topology to amplify the effects of weakly favored mutations under directed evolution in industrial applications.