The ability to design proteins from scratch, known as de novo protein design, opens new avenues for creating novel
biomolecules with desired functions, offering unparalleled potential in therapeutics, biocatalysis, and synthetic
biology. At CD ComputaBio, we pair cutting-edge computational methods with experimental validation to enable the
precision design and engineering of tailor-made proteins. Our state-of-the-art de novo protein design services are
tailored to meet your unique needs, whether you're looking to create enzymes, antibodies, peptides, or membrane
proteins.
Backgroud
De novo protein design is a multidisciplinary field combining principles from biophysics, computational biology, and
experimental molecular biology. Historically, protein engineering relied on modifying existing proteins. However, the
precision and versatility of de novo design allow for the construction of entirely new protein structures with bespoke
functions. At CD ComputaBio, our experts leverage advanced computational modeling tools and algorithms to simulate
protein folding and dynamics, followed by iterative design cycles to optimize protein stability and functionality.
Figure 1. De novo protein design.
Our Service
CD ComputaBio offers a comprehensive suite of de novo protein design services to cater to diverse research and
development objectives. Our services fall into the following main categories:
We design enzymes with enhanced catalytic activity, specificity, and stability for various industrial and
biomedical applications. Our team can create enzymes that can catalyze novel reactions or improve existing ones,
providing innovative solutions for synthetic chemistry and biotechnology.
Custom-designed antibodies with high affinity and specificity for targeted antigens. These antibodies can be
developed for diagnostic, therapeutic, and research purposes, offering new opportunities in the field of
immunology and disease treatment.
Designing peptides with specific biological activities, such as antimicrobial, anticancer, or immunomodulatory
effects. Peptides can be engineered to have optimal pharmacokinetic and pharmacodynamic properties for efficient
drug delivery and therapeutic efficacy.
Membrane proteins play crucial roles in various cellular processes and are often challenging to study and
engineer. Our services focus on designing membrane proteins with desired transport, signaling, or receptor
functions, contributing to advancements in membrane biology and drug discovery.
De Novo Binder Design with BindCraft
BindCraft is an open-source, automated pipeline for de novo protein binder design. It has been integrated into our self-developed protein design platform. CD ComputaBio can provide insights into writing and customizing loss functions to guide design objectives, along with practical ideas for the experimental validation of designed binders.
3D structure of the target protein (PDB format)
Define Binding Region (Optional but Recommended)
Set Up the Binder Scaffold
Configuration File (YAML/JSON format)
Given a target protein structure, a binder backbone and sequence is generated using AF2 multimer, then the surface and core of the binder are optimized using MPNNsol while keeping the interface intact. Finally, designs are filtered based on AF2 monomer model prediction.
Schematic representation of the BindCraft binder design pipeline. (Pacesa, M., et al. 2025)
PDB files → 3D models of target-binder complexes.
FASTA files → binder sequences.
Scoring reports → confidence scores, binding metrics, energy terms.
Ranked results → best candidates sorted by binding quality.
De novo design aims to create a novel molecule from the ground up, given an antigen and/or an epitope site to be targeted. (Bielska W, et al. 2025)
Integrates AF2 Multimer, ProteinMPNN, and PyRosetta
High binder success rates (10-100%)
Works in a one-shot or low-shot setting
Open-source, user-friendly pipeline
Applications
De novo protein design has diverse applications across multiple industries and research areas:
Therapeutics: Custom-designed enzymes and antibodies for targeted drug delivery, cancer therapy, and treatment of
genetic disorders.
Industrial Biotechnology: Engineering novel enzymes for efficient biocatalysis in chemical, food, and biofuel
industries.
Synthetic Biology: Constructing synthetic pathways and functional biomolecules for bioengineering and metabolic
engineering.
Our Algorithm
Sequence Prediction
By using a combination of deep learning and evolutionary algorithms, we predict amino acid sequences that are
likely to fold into the desired tertiary structure. Our models are trained on vast datasets, ensuring high
accuracy and robustness in sequence prediction.
Structure Prediction
We employ sophisticated molecular dynamics simulations and energy minimization techniques to predict the most
stable three-dimensional structures of the designed proteins. Our approach ensures that the protein will
maintain its desired configuration under physiological conditions.
Functionality Evaluation
Our service assess the potential functionality of the designed proteins by simulating interactions with other
molecules. This includes binding affinity calculations for target molecules, ensuring that the designed protein
folds correctly.
Sample Requirements
To initiate a de novo protein design project, we require specific information and samples from our clients:
Target Information: Detailed description of the target protein or desired function, including any available
structural data, sequence information, and biological context.
Application Details: Intended application of the designed protein, including any specific performance criteria or
constraints.
Material Samples: Where applicable, provision of starting materials such as plasmid DNA, expression systems, or
target ligands.
Experimental Conditions: Information on the intended experimental conditions, including temperature, pH, and any
relevant cofactors or substrates.
Results Delivery
CD ComputaBio is committed to providing timely and comprehensive results to our clients. Our results delivery process
includes:
Design Reports: Detailed reports summarizing the computational design process, including structural models, design
rationales, and predicted properties.
Experimental Data: Rigorous experimental validation data, including activity assays, binding studies, and
stability tests.
Final Constructs: Delivery of engineered protein constructs, including plasmids, purified proteins, or synthetic
peptides, as per project requirements.
Continuous Support: Ongoing support for data interpretation, troubleshooting, and potential iterations of the
design process.
Our Advantages
Expertise
Our team comprises seasoned experts in computational biology, structural biology, and protein engineering.
Advanced Tools
We utilize state-of-the-art computational tools and algorithms for accurate and efficient protein design.
Integrated Workflow
Seamless integration of computational design and experimental validation ensures robust and reliable results.
De novo protein design represents a transformative approach to engineering novel proteins with tailored functions,
offering limitless possibilities in therapeutics, industrial applications, and synthetic biology. CD ComputaBio is
your trusted partner in this cutting-edge field, providing a comprehensive suite of de novo design services, from
enzymes and antibodies to peptides and membrane proteins.
Frequently Asked Questions
How does De novo Protein Design differ from traditional protein engineering?
De novo Protein Design differs from traditional protein engineering primarily in its approach and objectives.
Traditional protein engineering often involves modifying existing proteins to enhance their properties or impart
new functions. This is typically achieved through techniques such as site-directed mutagenesis, where specific
amino acids in a known protein structure are altered to achieve desired characteristics.
In contrast, De novo Protein Design involves designing entirely new protein sequences and structures without
reference to any existing proteins. This requires a deeper understanding of how sequences relate to structures
and functions, as well as sophisticated computational tools to model and predict these relationships. While
traditional protein engineering relies on a "library" of existing proteins, De novo Protein Design
opens up a new "library" composed of infinite possibilities.
What are the key techniques used in De novo Protein Design?
Several key techniques are employed in De novo Protein Design, often involving a combination of computational
algorithms and experimental validation:
Computational Modeling: Algorithms like molecular dynamics simulations and energy minimization are used to
predict how a particular amino acid sequence will fold into three-dimensional structures.
Structure Prediction: Tools such as AlphaFold and Rosetta are employed to predict protein structures based on
their amino acid sequences. These tools utilize deep learning and physics-based modeling.
Sequence Design: Computational methods like evolutionary algorithms and Monte Carlo simulations help design
sequences that will fold into the desired structure.
Generating Libraries: High-throughput screening techniques may be used to create vast libraries of designed
proteins, which can then be tested for desired characteristics.
What are the main applications of De novo Protein Design?
De novo Protein Design has numerous applications across various fields, including but not limited to:
Pharmaceuticals: Designing novel enzymes or antibodies that can serve as therapeutic agents
against diseases.
Biotechnology: Creating enzymes with enhanced stability or specificity for industrial
applications, like biofuels or biodegradable plastics production.
Synthetic Biology: Engineering proteins that serve as molecular machines or tools for gene
editing, such as CRISPR-associated proteins.
Diagnostics: Developing proteins that can act as biosensors for detecting pathogens or
environmental pollutants.
Agriculture: Designing proteins that can enhance crop resistance to diseases or improve
nutrient uptake.
What challenges are associated with De novo Protein Design?
While exciting, De novo Protein Design faces several challenges:
Computational Complexity: The vast number of possible sequences and structures makes it
computationally demanding to predict which will achieve the desired outcome.
Experimental Validation: Synthesizing designed proteins and verifying their functions can
be resource-intensive and time-consuming.
Folding and Stability: Predicting how a protein will fold and ensuring it remains stable in
various environments is notoriously difficult.
Function Prediction: Accurately predicting the biological function of a designed protein
can be challenging, especially when designing entirely novel proteins.
References
1.Pacesa, M., Nickel, L., Schellhaas, C. et al. One-shot design of functional protein binders with BindCraft. Nature (2025).
Bielska W, Jaszczyszyn I, Dudzic P, Janusz B, Chomicz D, Wrobel S, Greiff V, Feehan R, Adolf-Bryfogle J and Krawczyk K. Applying computational protein design to therapeutic antibody discovery - current state and perspectives. Front. Immunol (2025).
3.Zheng J, Wang Y, Liang Q, Cui L, Wang L. The Application of Machine Learning on Antibody Discovery and Optimization. Molecules (2024).
For research use only. Not intended for any clinical use.
Figure 1. De novo protein design.
