Research
We develop computational and experimental technologies for designing molecular recognition, stabilizing viral antigens, controlling protein assembly, and learning from high-throughput biological measurements.
We develop generative and predictive approaches for protein interfaces, antibodies, nanobodies, and miniproteins. Our goal is to connect model development directly to experimental outcomes rather than relying only on retrospective structural benchmarks.
We redesign viral surface proteins to preserve vulnerable conformations, display conserved determinants, and test how sequence and geometry shape the breadth and quality of antibody responses.
We create de novo scaffolds, trimers, nanoparticles, and higher-order assemblies with defined geometry. These systems can organize antigens, tune receptor engagement, and serve as programmable biomolecular materials.
We use pooled oligonucleotide synthesis, display technologies, pseudoviruses, sorting, and NGS to measure large design spaces and build closed Design–Build–Test–Learn cycles.
Biological focus
Viral proteins solve difficult problems in recognition, membrane fusion, assembly, and delivery. Understanding those mechanisms allows us to inhibit them—or repurpose their underlying principles.
Immunodominance, immune imprinting, HA engineering, and high-resolution serology.
Structure-guided stabilization and immunogen design for F and HN proteins.
Designed scaffolds and assemblies for native-like antigen presentation.
Prediction, generative design, screening, and experimentally grounded learning.
Protein systems that organize, target, or deliver biological function.
Featured platform
Explore our open-source framework for antibodies, nanobodies, and miniproteins.
Design–Build–Test–Learn
ASCENT turns high-throughput measurements into reusable data, benchmarks, and improved models.