Our technology platforms
Integrated computational and experimental platforms for protein engineering, cell display screening, and AI-driven protein design
AI-driven protein design and binder discovery
Generative AI models for de novo protein design. No API wrappers. No shared batch queues.
RFdiffusion
Structure-guided diffusion model generating de novo protein backbones conditioned on target hotspots. Backbone diversity is controlled via noise schedule and contig specification, enabling anything from unconditional monomer generation to hotspot-guided binder design.
Learn more →BindCraft
Integrated binder design pipeline combining scaffold generation with iterative sequence optimization. BindCraft iterates between backbone generation, sequence design, and AlphaFold2 validation in a single loop, producing candidates that pass structural confidence filters before experimental screening.
Learn more →Boltzgen
Flow-based generative model for protein conformational sampling. Explores alternative backbone topologies to diversify the candidate pool across a broader region of fold space.
Learn more →Epitope Scout
Automated surface epitope identification and scoring for de novo binder design. Upload a PDB, get ranked epitope patches. Free to use.
Try Epitope Scout →ProteinMPNN
Message-passing neural network solving the inverse folding problem. Designs amino acid sequences that fold into generated backbones from RFdiffusion and Boltzgen.
Learn more →Structural validation
Boltz-2, ESMFold, and ColabFold (AlphaFold2) provide layered fold and complex validation. Confidence metrics filter candidates before synthesis.
Learn more →High-throughput display and characterization
Computationally designed candidates are validated experimentally using the same platforms we operate for traditional library campaigns.
Yeast Surface Display
High-throughput screening of protein variants on yeast cell surfaces. FACS and MACS selection with tunable stringency across 2-4 sorting rounds. Our most established platform for antibody discovery, affinity maturation, and de novo binder validation.
Learn more →Mammalian Display
Display on mammalian cells for targets requiring native post-translational modifications, proper glycosylation, or mammalian folding chaperones. Candidates that pass mammalian display are more likely to retain function during downstream production, reducing late-stage attrition.
Learn more →Deep Mutational Scanning
Systematic measurement of variant fitness across thousands of single and combinatorial mutations. Comprehensive fitness landscapes for stability, expression, and function in a single experiment. Also applied to drug target biology, mapping how mutations in a target protein affect drug binding, resistance, and mechanism of action.
Learn more →Directed Evolution
Iterative rounds of random and focused mutagenesis followed by high-throughput selection. Evolve proteins for stability, activity, specificity, or expression without requiring structural information.
Learn more →Variant Library Construction
Custom combinatorial and saturation mutagenesis libraries. Error-prone PCR, NNK scanning, and computationally designed focused libraries cloned into display or expression vectors.
Learn more →Small-Scale Expression & BLI
Top hits from display screening are expressed as soluble protein in small-scale cultures and characterized by biolayer interferometry. BLI provides real-time binding kinetics (kon, koff, KD) to confirm target engagement and rank candidates before scale-up.
Learn more →NGS Analysis Pipeline
Next-generation sequencing pipeline for quantitative, reproducible hit calling and ranking. Enrichment scoring, clone frequency tracking across sorting rounds, and prioritized hit lists with statistical confidence metrics.
Learn more →From target structure to validated hits
Computational design generates candidates. Experimental screening validates them. NGS analysis closes the loop. Every stage feeds data forward.
Ready to leverage our technology platforms?
Tell us about your protein engineering challenge. We will assess which platforms are the right fit and propose a program structure.
Start a project →