Designing DARPins with NeuroBind’s New Affinity Maturation Feature

Written by Keaun Amani

Published 2025-7-28

Introduction to NeuroBind

In this post, we'll walk through how to design high-affinity DARPins using NeuroBind’s cutting-edge affinity maturation feature. If you're new to DARPins, check out our previous post: DARPins vs. Antibodies: A Comprehensive Guide to Next‑Gen Protein Scaffolds.

NeuroBind is a powerful in silico protein binder design algorithm capable of both de novo design and affinity maturation. The latest update enables users to input existing binders—whether antibodies, nanobodies, peptides, or DARPins—and automatically optimize them for improved binding, stability, solubility, and reduced immunogenicity.

Optimize and perform your own affinity maturation experiments 🔗 online using NeuroBind.


Step 1: Preparing the DARPin Template

1. Choose a structural template.

We’ll use PDB 5OOU, a DARPin scaffold with well-defined ankyrin repeats as our starting point.

2. Define the target: PD‑L1.

Program your design against Programmed Death Ligand 1 (PD‑L1), a critical immune checkpoint protein involved in suppressing anti-tumor immunity.

3. Upload to NeuroBind.

Under the Custom Template option, upload 5OOU.pdb for design refinement.

4. Include only structured domains.

When providing both the Target Sequence and Custom Template, it is generally advisable to exclude disordered regions—particularly those commonly found at the N- and C-termini. While not strictly required, retaining these flexible segments can increase computational cost and lead to designs that engage peripheral or non-functional surfaces of the target. In this experiment, we truncate the PD‑L1 sequence to include only the structured, functionally relevant domains, ensuring that NeuroBind focuses its design efforts on the most meaningful regions for high-affinity binding.

5. Skip manual paratope/epitope mapping.

NeuroBind can automatically determine the optimal binding interface, saving you time and expertise.


Step 2: Submitting the Design Job

Once your DARPin scaffold and PD‑L1 target are ready:

NeuroBind typically returns 25 refined candidates, ranked by overall quality.


Step 3: Interpreting NeuroBind Results

NeuroBind’s result table contains:

Metric Definition
Rank Position among candidates, sorted by overall quality
Overall Quality Score (0–1); higher means better predicted performance
Sequence Amino acid sequence of the DARPin binder
Affinity Predicted binding free energy (kcal/mol); more negative values ≈ stronger bind
Thermostability Predicted thermal resilience (0–1); higher = better stability at high temps
Solubility Likelihood of expression in solution (0–1); higher = better solubility
Immunogenicity Predicted risk of immune response (0–1); higher = lower immunogenicity (antibody/scFv modes only)
Hydrogen Bonds Number of predicted H‑bonds at the interface

NeuroBind also provides a 2D scatter plot to visualize structurual similarity and clustering among designs, aiding in candidate selection.

Top binder design against PD‑L1 from NeuroBind


Step 4: Choosing the Best Candidate 🧬

To identify the top binder for PD‑L1:

  1. Look for high Overall Quality—ideally > 0.8.
  2. Target Affinity ≤ –15 kcal/mol indicates strong binding.
  3. Prioritize thermostability & solubility > 0.7.
  4. Minimize immunogenicity—aim > 0.8.
  5. Consider interface hydrogen bonding, supporting specificity.

Select a candidate that balances these properties for experimental validation or further engineering.

NeuroBind results table + scatter plot for our PD‑L1 binders


Conclusion

With NeuroBind’s new affinity maturation feature, designing optimized DARPins against therapeutic targets like PD‑L1 is faster and smarter than ever. By taking a robust scaffold like PDB 5OOU and refining it automatically, users benefit from:

👉 Ready to create optimized binders? Try NeuroBind here.

Explore more posts

Conformational Pre-Organization: The Silent Key To Effective Binder Design

By Keaun Amani

DARPins vs. Antibodies: A Comprehensive Guide to Next-Gen Protein Scaffolds

By Keaun Amani

Exploring the Power of Protein-DNA and Protein-RNA Docking

By Amélie Lagacé-O'Connor

Protein-Protein Docking Simplified: Illuminating the Mechanics of Protein Interactions

By Amélie Lagacé-O'Connor

Binding Affinity, Gibbs Free Energy, And Intramolecular Energy Explained for Drug Discovery

By Amélie Lagacé-O'Connor

Understanding the Differences between AI, Machine Learning, and Deep Learning

By Keaun Amani

Accelerate your lab's
research today

Register for free — upgrade anytime.

Interested in getting a license? Contact Sales.

Sign up free