Ref K1
Computational Service
Confidence Reported
Scoped After Feasibility Review
Service Overview
De novo design builds proteins that have no natural counterpart: a backbone is generated to satisfy a geometric or functional requirement, a sequence is designed to fold into it, and the design is then filtered computationally before anything is expressed. The approach now produces binders, scaffolds and minimal folds at usable success rates, but those rates are still measured in percent of designs tested. We therefore design panels, filter them aggressively, and are explicit about the numbers involved.
Key Features & Advantages
Backbone Generation Then Sequence Design
Structure is generated first and sequence designed onto it, which produces folds matched to the functional requirement rather than borrowed from a natural protein.
In Silico Filtering Before Expression
Designs are filtered by structure prediction self-consistency, predicted confidence, interface metrics and stability, which is what raises experimental success rates from negligible to workable.
Motif Scaffolding Supported
A functional motif — an epitope, a binding loop, a catalytic arrangement — can be held fixed while a new scaffold is built around it to present it correctly.
Expression Realism Considered
Designs are screened for aggregation propensity, disulfide handling and expression-relevant features, since a design that cannot be produced is of no use.
Where This Service Is Used
1Designing novel binders against a defined target surface
2Building scaffolds that present an epitope correctly for vaccine or reagent applications
3Creating miniproteins as alternatives to antibodies for imaging or delivery
4Designing symmetric assemblies, nanocages or oligomeric scaffolds
5Generating stable scaffolds to carry a functional motif that is unstable in isolation
Methods and Techniques Applied
1Definition of the design objective: target surface, motif geometry, symmetry or fold requirement
2Backbone generation with diffusion or hallucination methods under the stated constraints
3Sequence design onto the generated backbones with fixed-backbone design methods
4Self-consistency filtering: predicting the structure of each designed sequence and comparing it with the intended backbone
5Interface, stability and packing assessment, with molecular dynamics for the leading designs
6Expression-risk screening and selection of the panel to be synthesised and tested
What We Need From You
1Target structure and the surface to be bound, or the motif and geometry to be scaffolded
2Functional requirements: affinity, specificity, stability, size limits, oligomeric state
3Expression system available, since it constrains disulfides, glycosylation and construct design
4The number of designs you can express and screen, which determines panel size
What You Receive
1Designed sequences with structural models and per-design filtering metrics
2Self-consistency and confidence analysis for every delivered design
3Interface analysis for binder designs, showing the intended interactions
4Expression construct suggestions including tags, linkers and boundaries
5Ranked testing order with a realistic statement of expected success rates
What We Compute and Report
1Designed sequences with predicted structures and per-design filtering metrics
2Self-consistency between the designed sequence and the intended backbone, in ångströms
3Interface metrics for binder designs: buried area, shape complementarity, predicted contacts
4Predicted stability and aggregation risk per design
5Expression-risk indicators used to order the panel for testing
Questions We Are Asked Before Scoping
What success rate should we expect?
▸For binder design against a defined surface, published campaigns often report a small percentage of tested designs showing measurable binding. Panels must be sized with that in mind, and we say so before the project starts.
Can you design an enzyme from scratch?
▸De novo catalysis remains substantially harder than binding. We can scaffold a catalytic motif and design the surrounding fold, but expectations should be set well below those for binder design.
Do designs express well?
▸Designed proteins are frequently more stable and better expressed than natural ones, but failures occur. We screen for expression risk and recommend testing several constructs per design.
Interpretation and Limitations
Computational predictions prioritise hypotheses; they do not independently establish binding, biological activity, therapeutic efficacy, safety or clinical validity. Performance depends on input quality, target class, the structural evidence available and the applicability domain of the selected methods. Scores used for ranking are comparative devices and should not be read as measured quantities. All deliverables are for research use only and are not intended for diagnostic or therapeutic use.
Is This the Right Service for You?
Choose this when
A small, stable binder is wanted as an alternative to an antibody
An epitope or functional motif must be presented on a purpose-built scaffold
Screening capacity exists for a designed panel rather than a single construct
Consider another service when
An existing protein needs stabilising rather than replacing
Protein Stability & Thermostability Engineering An antibody format is acceptable and more precedented
Full-Length IgG Antibody Design What Goes Wrong, and How We Handle It
These are the failure modes that produce confident-looking results which do not survive the next experiment. Each one is designed against rather than discovered afterwards.
Panels sized on best-case published success rates
Expected hit rates are stated before the project starts and panel size is set from them, not from the most favourable paper.
Designs that model beautifully and never express
Aggregation propensity, disulfide handling and expression-relevant features are screened, and construct variants are recommended.
Sequence and intended structure quietly disagreeing
Self-consistency filtering predicts the structure of each designed sequence and compares it against the intended backbone before delivery.
A Worked Example
Miniprotein binders against a defined surface patch
The situation
The team needed a small binder for an imaging application where an antibody was too large.
What the work involved
Backbones were generated against the target patch, sequences were designed onto them, self-consistency and interface metrics were used to filter aggressively, and the survivors were simulated briefly.
What came back
Ninety-six designs selected from several thousand generated, delivered with per-design metrics and expression construct suggestions, and with the expected success rate stated up front.
Illustrative of the shape this work takes. Details are generalised, no client or target is identified, and scope and outcome vary with the system in question.
What Published Work Reports
The results below were published by independent groups and are cited to show what these computational approaches have achieved in the literature. They are not ICDMO project results, and outcomes vary substantially between targets, datasets and programmes. We do not promise comparable results for any particular project.
Machine-learning-guided evolution raised a reaction yield from 12% to 93%
An active-learning workflow optimising five interacting active-site residues improved the yield of a desired non-native cyclopropanation product from 12% to 93% in three rounds of laboratory experimentation. The design principle is the one we apply: computation chooses which small set of variants to make, and measured results retrain the model between rounds.
Nature Communications, 2025 ↗1.6- to 42-fold activity improvements from model-selected enzyme variants
A machine-learning-guided enzyme engineering campaign evaluated 1,217 variants across 10,953 reactions to build predictive models for amide synthetases. Model-selected variants showed 1.6- to 42-fold improved activity relative to the parent enzyme across nine pharmaceutical targets, illustrating both the scale of data that makes models predictive and the spread of outcomes that remains.
Nature Communications, 2024 ↗De novo binder design succeeds at rates that vary enormously by target
Recent design methods report nanomolar binder hit rates between roughly 17% and 82% across evaluated targets, while independent evaluations of earlier pipelines on other targets report designs failing on expression, non-specific binding or undetectable affinity. The honest reading is that de novo binder design now works, that success is strongly target-dependent, and that panels must be sized accordingly rather than on a best-case figure.
Comparative design studies, bioRxiv, 2025 ↗Service Workflow
1
Scientific Consultation
We discuss the objective, the decision it supports and the data you hold. Nothing is quoted before we understand what the result has to answer.
2
Feasibility Review
We check that the data, target class and available structural evidence can support the proposed approach, and say plainly where they cannot.
3
Technical Proposal
Scope, methods, deliverables, timeline and price are confirmed in writing, together with the limitations we expect to report.
4
Computation and QC
The analysis runs on pinned tool versions in a containerised environment, with quality checkpoints and method comparison where that is informative.
5
Report and Follow-Up
You receive the ranked results, the documented methods, the limitations and a proposed validation plan, plus support in interpreting them.
Service Process
1
Online Consultation
›
2
Solution Matching
›
3
Service Contract
›
4
Project Execution
›
5
Data Delivery
Specifications & Turnaround
Note: ICDMO delivers computational (in silico) services. Results prioritise hypotheses for testing; they do not by themselves establish binding, biological activity, efficacy or safety. All deliverables are for research use only and are not intended for diagnostic or therapeutic use. Downstream experimental work is scoped separately and executed through qualified partners.
Get a Custom Quote
Our scientific team responds within 24 hours with a detailed technical proposal and pricing tailored to your project.
Contact Us Online ConsultationStandard Deliverables
✓Sequence-verified final product
✓Full QC documentation
✓Certificate of Analysis (CoA)
✓Electronic data files
✓Project completion report
Why ICDMO?
Computational CDMO Focus
Reproducible, Versioned Pipelines
Confidence & Limitations Reported
Experiment-Aware Recommendations
Response Within One Business Day