We build novel protein structures from scratch using state-of-the-art diffusion models and deep learning. Our integrated computational-experimental pipeline delivers designs with validated structural and functional metrics.
The sequence space for a 100-residue protein contains roughly 20100 (~10130) possibilities, yet natural evolution has sampled only a vanishingly small subset. De novo design allows us to access this vast unexplored territory.
Create proteins with structures and functions absent from the natural repertoire. Access binding sites and catalytic activities that evolved systems have not discovered.
Define secondary structure composition, active site geometry, oligomeric state, and binding interface architecture computationally, then generate matching scaffolds.
Generate hundreds of backbone candidates rapidly. Move from computational design to expression-ready sequences efficiently.
Start with a target function and derive structure. Scaffold catalytic residues, engineer allosteric regulation, and build biosensors from physical principles.
Enginoma integrates deeply re-engineered computational architectures into a validated workflow, from backbone generation through structural confirmation, calibrated on proprietary wet-lab datasets.
Enginoma Backbone generates novel protein scaffolds through a proprietary diffusion-based process, performance-calibrated on curated structural benchmarks and proprietary validation data.
Message Passing Neural Network for protein sequence design. Given a target backbone, it generates amino acid sequences with high predicted foldability. Published recovery rates of ~52% substantially exceed Rosetta-based approaches.
Designed sequences are validated through Enginoma Structure refolding with comprehensive confidence metrics. High confidence scores and low self-consistency RMSD indicate designs likely to fold as intended.
Four design modes covering the majority of protein engineering applications
Generate novel single-chain proteins
Starting from noise, we generate diverse backbones spanning all major topologies: alpha-helical, beta-sheet, and mixed folds. Validated designs up to 600 amino acids with matched secondary structure composition.
Specify fold architecture
Guide generation toward specific secondary structure patterns: helical bundles, beta-barrels with defined pore sizes, or TIM-barrel folds. Control global architecture while allowing local sequence diversity.
Engineer multimeric assemblies
Design homo-oligomeric assemblies with precise symmetry (C2 through C12). Applications include vaccine scaffolds with defined antigen valency and multivalent binding platforms.
Preserve functional sites in new frameworks
Embed known functional motifs—catalytic triads, metal coordination spheres, or binding epitopes—within novel scaffolds. Optimize stability, expression, and immunogenic properties while maintaining precise functional geometry.
Integrated computational design followed by experimental characterization at each stage
We define design objectives with you: protein size, fold class, functional requirements, symmetry constraints, and specific structural features that establish success criteria.
Enginoma Backbone produces 50-200 diverse scaffold candidates matching your specifications. We rank by structural quality, diversity coverage, and predicted stability.
Enginoma Sequence designs amino acid sequences for each backbone. Enginoma Structure validates refolding. We filter for high confidence scores and low self-consistency RMSD.
Top candidates move to the wet lab: gene synthesis, E. coli expression, purification, and biophysical characterization (CD, SEC-MALS, thermal stability).
De novo designed proteins are advancing research across multiple sectors
Novel binding proteins, cytokine mimetics, and antibody alternatives with optimized specificity, potency, and developability for challenging targets.
Novel enzyme scaffolds with tailored substrate specificity and catalytic efficiency. Design catalytic sites from first principles for non-natural transformations.
Self-assembling protein nanoparticles displaying antigenic epitopes with defined valency and geometry for enhanced immune responses.
Allosteric switches and ligand-responsive proteins for diagnostic assays, environmental monitoring, and synthetic biology circuits.
Symmetric protein cages, filaments, and 2D lattices for targeted drug delivery, bioactive materials, and structural biology tools.
Fluorescent reporters, optogenetic actuators, and affinity reagents with customized binding properties for basic research.
Our pipeline builds on peer-reviewed methods published in leading journals
Watson, J.L. et al. De novo design of protein structure and function with Enginoma Backbone. Nature 620, 1089-1100 (2023). https://doi.org/10.1038/s41586-023-06415-8
Enginoma Backbone: structure-conditioned protein generation using diffusion models.Dauparas, J. et al. Robust deep learning-based protein sequence design using Enginoma Sequence. Science 378, 49-56 (2022). https://doi.org/10.1126/science.add2187
Enginoma Sequence: message passing neural network for sequence design.Abramson, J. et al. Accurate structure prediction of biomolecular interactions with Enginoma Structure 3. Nature 630, 493-500 (2024). https://doi.org/10.1038/s41586-024-07487-w
Enginoma Complex: unified structure prediction including protein complexes.Krishna, R. et al. Generalized biomolecular modeling and design with Enginoma Structure All-Atom. Science 384, eadl2528 (2024). https://doi.org/10.1126/science.adl2528
Enginoma Structure All-Atom: all-atom modeling of proteins and small molecules.Bennett, N.R. et al. Atomically accurate de novo design of single-domain antibodies. Nature 637, 339-347 (2025). https://doi.org/10.1038/s41586-025-09721-5
De novo design of picomolar-affinity single-domain antibodies using Enginoma Backbone.Our platform handles proteins from 50 to 600+ amino acids. Smaller designs (50-150 aa) suit binding domains, while larger proteins (200-600 aa) enable complex enzyme scaffolds and multimeric assemblies.
Project timelines vary based on protein complexity and design requirements. Contact us for a detailed project assessment and timeline estimate.
Validation success rates vary depending on protein complexity and design objectives. We provide detailed risk assessment and validation strategies for each project.
Enginoma Backbone for scaffold generation, Enginoma Sequence for sequence design, and Enginoma Structure for structural validation—continuously improved through closed-loop wet-lab feedback.
We offer CD spectroscopy for secondary structure, SEC-MALS for oligomeric state, and SAXS for solution structure. High-resolution validation by X-ray crystallography or cryo-EM is available through partner facilities.
Whether you need a binding protein, enzyme scaffold, or entirely new fold, our team can guide you from concept to validated design.
Tell us about your project and we'll get back within 24 hours.
