A comprehensive approach to prediction of stabilizing mutations in G-protein coupled receptors yields high hit rate and crystal structures of 5HT2C in both active and inactive states.

Six computationally designed and in vitro optimized protein inhibitors serve as molecular probes to reveal BCL2 profiles of different human cancers.

The first successful de novo design of a homo-trimeric protein that binds a C3 symmetric small molecule larger than a metal ion is an advance for computational protein design.

Protein structures with complex topologies can be designed by recombining small independently folded domains.

State-of-the-art computational methods have facilitated the construction of plant-based sentinels for detecting toxic molecules in the environment.

The combination of computational modeling and protein design can reveal key determinants of antibody–antigen binding and optimize small sets of antigen variants for efficient experimental localization of epitopes.

Key allosteric residues are predicted based on sequence coevolution analysis, which serves as guidelines for allosteric drug design and optimization.

Short peptides that bind tightly to anti-apoptotic protein Bfl-1 but not other Bcl-2 family members provide a tool for diagnosing cancer cell survival mechanisms and a lead for developing new therapeutics.

A citizen science game reveals play interest patterns that could inform crowdsourcing experiment design.

Designer transmembrane peptides precisely control chimeric antigen receptor signaling strength and are insulated from undesired interactions with endogenous receptors.