Background: Protein aggregation into β-sheet-enriched insoluble assemblies is being
found to be associated with an increasing number of debilitating human pathologies, such as Alzheimer’s
disease or type 2 diabetes, but also with premature aging. Furthermore, protein aggregation
represents a major bottleneck in the production and marketing of proteinbased therapeutics.
Thus, the development of methods to accurately forecast the aggregation propensity of a certain
protein is of much value.
Methods/Results: A myriad of in vitro and in vivo aggregation studies have shown that the aggregation
propensity of a certain polypeptide sequence is highly dependent on its intrinsic properties
and, in most cases, driven by specific short regions of high aggregation propensity. These observations
have fostered the development of a first generation of algorithms aimed to predict protein
aggregation propensities from the protein sequence. A second generation of programs able to map
protein aggregation on protein structures is emerging. Herein, we review the most representative
online accessible predictive tools, emphasizing their main distinctive features and the range of
Conclusion: In this review, we describe representative biocomputational approaches to evaluate
the aggregation properties of protein sequences and structures, while illustrating how they can
become very useful tools to target protein aggregation in biomedicine and biotechnology.